THE EFFECTS OF INTERNET ADDICTION ON COLLEGE ..., Study notes of History of Science and Technology

Download THE EFFECTS OF INTERNET ADDICTION ON COLLEGE ... and more Study notes History of Science and Technology in PDF only on Docsity! THE EFFECTS OF INTERNET ADDICTION ON COLLEGE STUDENTS: THE RELATIONSHIP BETWEEN INTERNET ADDICTION TEST SCORES, COLLEGE STUDENT DEMOGRAPHICS, AND ACADEMIC ACHIEVEMENT By MICHAEL L. HOUSTON Bachelor of Science in Human Relations Southern Nazarene University Bethany, Oklahoma 1998 Master of Education in College Student Affairs Azusa Pacific University Azusa, California 2000 Submitted to the Faculty of the Graduate College of the Oklahoma State University in partial fulfillment of the requirements for the Degree of DOCTOR OF PHILOSOPHY December, 2019 ii THE EFFECTS OF INTERNET ADDICTION ON COLLEGE STUDENTS: THE RELATIONSHIP BETWEEN INTERNET ADDICTION TEST SCORES, COLLEGE STUDENT DEMOGRAPHICS, AND ACADEMIC ACHIEVEMENT Dissertation Approved: Dr. Kerri Kearney ________________________________________________ Dissertation Adviser Dr. Stephen Wanger _____________________________________________________ Dr. Jane Vogler Cragun _____________________________________________________ Dr. Mwarumba Mwavita _____________________________________________________ v Couple that with your willingness to step into my moment of crazy and wear the advisor hat, and my debt to you is large. You are a true blessing to me, and words cannot express my gratitude for all you have done. Thank you so much, and WE DID THIS! ☺ -To the amazing faculty at OSU, you challenged me with new perspectives, and helped me become a better writer, scholar, and practitioner. I will always be indebted to you for the time and wisdom you gave and hope to continue in your scholarly footsteps. Dr. Z, you showed me how a true scholar’s mind works, and I am grateful for your feedback and the countless resources you provided. You were a springboard for this topic and project, and I hope our paths cross again. Dr. Bailey, the zest and fervor for life you brought each and every night of class was a gift! You brightened my life and I miss you. Know you are making a difference in this world, and I am just one example. -Dr. Foubert, when I first heard you at the ACSD at Taylor, I knew our paths would cross again. Who knew that you would be my doctoral advisor? You were an amazing advisor in every sense of the word. You championed me, pushed me, provided timely feedback and life insight. I truly am sad that we could not finish together, but your fingerprints are all over this work. Thank you for giving of yourself! A friend told me this was the most difficult section to write. I am now beginning to realize why. There is not enough room to continue to thank the countless people along this doctoral journey. So many have come alongside and provided support and insights. Dr. Drabenstot thank you for fielding my stats questions and verbally processing my alignment and questions. Mary Eskridge thank you for proofreading this beast of a project. Scott and Mike thank you for giving me the time to work on this degree and helping financially. Chuck if you know you know. Kris your friendship is special, and I am so blessed to have it, and to the rest of my family and friends who have supported prayerfully and financially, I complete this for you. -Finally, “To Him who is able to do exceedingly abundantly above all that we ask or think, according to the power that works in us, to Him be glory forever and ever.” Ephesians 3: 20-21. *Acknowlegements reflect the views of the author and are not endorsed by committee members or Oklahoma State University. vi Name: MICHAEL L. HOUSTON Date of Degree: DECEMBER, 2019 Title of Study: THE EFFECTS OF INTERNET ADDICTION ON COLLEGE STUDENTS: THE RELATIONSHIP BETWEEN INTERNET ADDICTION TEST SCORES, COLLEGE STUDENT DEMOGRAPHICS, AND ACADEMIC ACHIEVEMENT Major Field: EDUCATIONAL LEADERSHIP AND POLICY STUDIES: HIGHER EDUCATION Abstract: Modern technology has changed education in many ways in a very short time. Not only are college students using technology daily, technology innovations help educators reach broader audiences of students through online learning, and online portals help educators share course materials. Awareness of this modern technology and the impacts it is having on higher education and students has become a critical issue over the past two decades. Even though the benefits of technology are often visible, researchers are finding that technology is creating challenges for some students. Students’ access to personal technologies has drastically increased, and with it the level of distraction, which competes with academic interests. The purpose of this quantitative study was to analyze the relationship between the Internet Addiction Test (IAT) score and academic performance. The IAT measured the student’s addiction to the Internet based upon his or her use. A student’s academic performance was measured by grade point average. A sample of 692 traditionally aged college students from both public and private institutions was used to examine if IAT scores were related to and predictive of grade point average. Data analysis comprised four stages: descriptive, correlation, linear regression, and analysis of variance. This study found a negative relationship was present between students’ IAT scores and grade point averages. While the statistics showed that as IAT scores increased, students’ grade point averages decreased, the overall affect was minimal. Better understanding of how Internet addiction is related to grade point average may prove helpful for higher education leaders. As technology innovations continue to rapidly increase, it is imperative that educators understand the relationship technologies have on college students. KEYWORDS: Internet addiction; Technology addiction; Internet Addiction Test vii TABLE OF CONTENTS Chapter Page I. INTRODUCTION ......................................................................................................1 Problem Statement ...................................................................................................3 Purpose .....................................................................................................................4 Research Questions and Hypothesis ........................................................................5 Design Overview .....................................................................................................6 Definition of Terms..................................................................................................7 Technology Addiction .................................................................................7 Problematic Internet Use..............................................................................7 Multitasking .................................................................................................8 Digital Natives .............................................................................................8 Cyberslacking ..............................................................................................9 Significance of Study ...............................................................................................9 Research .......................................................................................................9 Theory ........................................................................................................10 Practice .......................................................................................................11 Research Statement ................................................................................................12 Chapter Summary ..................................................................................................13 II. REVIEW OF LITERATURE..................................................................................15 College Student Demographics ..............................................................................17 Sex..............................................................................................................17 Classification .............................................................................................18 Race............................................................................................................18 Grade Point Average ..................................................................................19 Digital Natives .......................................................................................................20 Immediate Gratification for Effort .............................................................21 Desire for Speed and Frustration with Slow-paced Environments ............22 Challenged to Multitask .............................................................................22 More Pictures Less Text ............................................................................23 Nonlinear Information Processing .............................................................23 Collaborative Learning and Constant Connectivity ...................................24 Learning by Doing Rather than Lecture or Reading ..................................24 Balancing Time Between Work, Play, and Leisure ...................................25 Expectation for Technology in Educational Settings.................................25 x Chapter Page Race..........................................................................................................105 Research Site.. ..........................................................................................106 Implications..........................................................................................................107 Theory ......................................................................................................107 Rational Addiction Theory .............................................................107 Distraction Conflict Theory ............................................................108 Practice .....................................................................................................109 Identifying Students with Internet Addiction .................................110 Supporting Students with Internet Addiction .................................111 Using Technology Wisely in the Formal Academic Setting ..........112 Research ...................................................................................................113 Future Research ...................................................................................................114 Student Populations .................................................................................114 Types of Technology ...............................................................................115 Neurological Impacts of Technology Addiction......................................115 Different Methodologies ..........................................................................116 Qualitative Research Method ..........................................................117 Longitudinal Method ......................................................................117 Technology Addiction and Student Engagement ....................................118 Conclusion ...........................................................................................................119 REFERENCES ..........................................................................................................121 APPENDICES ...........................................................................................................144 APPENDIX A: Internet Addiction Test (IAT) Questions ...................................144 APPENDIX B: Email Template and IRB Consent Forms...................................149 APPENDIX C: Study Instrument ........................................................................152 APPENDIX D: Code Book ..................................................................................159 APPENDIX E: IRB Approvals ............................................................................161 APPENDIX F: Assumption Tables .....................................................................169 APPENDIX G: Statistical Data Tables ................................................................171 APPENDIX H: Data Figures ...............................................................................176 xi LIST OF TABLES Table Page 4.1. Descriptive Data of Study: Sex, Race and Classification ...............................82 4.2. Descriptive Data of Study: Research Site .......................................................83 4.3. Pearson correlation..........................................................................................84 4.4. Durbin-Watson test .........................................................................................87 4.5. Linear Regression Model Summary ...............................................................88 4.6. ANOVA ..........................................................................................................88 4.7. Coefficients .....................................................................................................89 4.8. Descriptives: Sex and IAT Score ....................................................................90 4.9. Test of Homogeneity of Variances: Sex and IAT Score.................................90 4.10. Robust Tests of Equality of Means: Sex and IAT Score ..............................90 4.11. Descriptives: Research Site and IAT Score ..................................................92 4.12. Test of Homogeneity of Variances: Research Site and IAT Score ...............92 4.13. Robust Tests of Equality of Means: Research Site and IAT Score ..............92 4.14. Descriptives: Race and IAT Score ................................................................93 4.15. Test of Homogeneity of Variances: Race and IAT Score .............................94 4.16. ANOVA: Race and IAT Score .....................................................................94 4.17. Descriptives: Classification and IAT Score ..................................................95 4.18. Test of Homogeneity of Variances: Classification and IAT Score...............95 4.19. Robust Tests of Equality of Means: Classification and IAT Score ..............96 xii LIST OF FIGURES Figure Page 4.1. Scatter Plot of GPA and IAT score .................................................................85 4.2. Residual Probability Plot ................................................................................86 4.3. Histogram of Residuals .................................................................................86 4.4. Scatterplot of residuals ....................................................................................87 4.5. Variance of Residuals .....................................................................................87 4.6. Normal Distribution of Residuals ...................................................................89 3 Problem Statement From books, to chalkboards, to television, to modern day tablet computers, smartphones and smart boards in the classrooms, technology innovations impact the higher education community (Haran, 2015). One visible change technology has on education is how instruction is provided to students. Online instruction has increased across higher education institutions at rapid rates and allows colleges and universities to reach populations of students unable to attend brick and mortar campuses (Kenney, 2011; Kurt, 2010; Lin & Yang, 2011; Mango, 2015). Additionally, technology innovations have impacted web portals, course management, and learning systems in higher education. Studies show that each of these innovations enhanced student learning and persistence (Christen, 2009; Costley, 2014; Keser, Uzunboylu & Ozdamli, 2012). Current literature also highlights the positive relationships between technology and student engagement, student confidence, and motivation (Costley, 2014; Kenney, 2011; Lin, & Yang, 2011). However, studies are also revealing negative effects of technology use on college students and challenging many of the positive presuppositions of educators regarding the use of technology in educational settings (Edwards, 2015; Fried, 2008). Compared to previous generations, college students today spend less time studying (Arum & Roska, 2011) in lieu of the many distractions vying for their time, and technology use is one of the most glaring. Indeed, many 21st century students are becoming addicted to technology (Agarwal & Kar, 2015; Young, 1998). Technology addiction is a psychological dependence on technology and is characterized by increased investment of time on technological pursuits (Nalwa & Anand, 2003; Young, 2008). College students 4 are entering institutions addicted to technology at rates that far outpace previous generations (Christakis et. al., 2011). Their addiction may inhibit their intended learning outcomes in higher education (Agarwal & Kar, 2015; Young, 1998). Students exhibiting signs of technology addiction show decreases in student success and persistence in higher education (Krumrei-Mancuso, Newton, Kim, & Wilcox, 2013). Additionally, a student’s use of personal computers, smartphones, and video games are linked with negative psychosocial behaviors which impact student learning (Heyoung, Heejune, Samwook & Wanbok, C., 2014; Hui-Jie, Hao-Rui & Wan-Seng, 2014; Schmitt & Livingston, 2015; Yao & Zhi-jin, 2014). Furthermore, neurology research is revealing relationships between extended technology use and negative impacts on brain chemistry and development (Liu et. al., 2015). In summary, scholars suggest that technology use may have negative impacts on student engagement, learning, and persistence in higher education today (Christakis et. al., 2011; Edwards, 2015; Fried, 2008). Further research is needed to determine how technology addiction is impacting college students’ academic performance along with the relationship of student demographics to technology addiction. Purpose The purpose of this quantitative study was to analyze the relationship of the Internet Addiction Test (IAT) score and academic performance and identify difference between student demographic variables and IAT scores. This study used Young’s (1998) Internet Addiction Test (IAT). The IAT score was used to determine a student’s addiction to the Internet. A student’s academic performance was measured by grade point average. A student’s demographics included sex, research site, race, and student classification. 5 Research Questions and Hypothesis Following the review of the literature and utilizing two theoretical frameworks (rational addiction theory and distraction conflict theory), the following research questions and hypotheses were created to guide this study: Research Questions RQ1: “Is there a relationship between a student’s Internet Addiction Test (IAT) score and grade point average?” RQ2: “Is there a mean difference among sex and IAT score?” RQ3: “Is there a mean difference among research site and IAT score?” RQ4: “Is there a mean difference among race and IAT score?” RQ5: “Is there a mean difference among classification and IAT score?” Hypotheses RQ1: H0/Null hypothesis: There is not a significant relationship between IAT score and grade point average. RQ1: Ha/Alternate hypothesis: Students’ scores on the IAT are significantly related with grade point average. RQ2: Ho/Null hypothesis: There is no significant difference of means of IAT score and sex. RQ2: Ha/Alternate hypothesis: There is a significant difference of means of IAT score and sex. RQ3: Ho/Null hypothesis: There is no significant difference of means of IAT score and research site. 8 Multitasking Multitasking refers to the brain switching back and forth between focal points or switching between multiple forms of information at one time (Junco & Cotten, 2012). Multitasking is a critical component of this research, as students are increasingly challenged to switch between nonacademic and academic tasks, oftentimes due to the available technology present (Ophir, Nass, & Wagner, 2009). Digital Natives The term many researchers use for 18-29 year olds, comprising today’s traditional college students, is digital natives (Bowe, & Wohn, 2015; Dede, 2004; Prensky, 2001; Tapscott, 2009; Thompson, 2013). Digital natives grew up with computers in the home and in school and had various forms of technology at their disposal. Additionally, the smart phone and social media were introduced when they were very young (Christakis et. al., 2011). This saturation of technology throughout a younger person’s life, both socially and academically for example, is a marker of this generation (Rideout, Foehr, & Roberts, 2010; Thompson, 2013). Some scholars question the desire digital natives have for lives completely enmeshed with technology (Friedl & Vercic, 2011). Furthermore, studies are trying to determine if there is a significant difference in learning preferences between digital natives and previous generations (Bowe & Wohn, 2015). Even with these disparate examples, the body of literature pertaining to digital natives is continuing to grow and show that this generation of college students is influenced by technology in the way they learn, process information, and interact socially (Barak & Dori, 2009; Tapscott, 2009; Thompson, 2013). 9 Cyber Slacking Cyber slacking is a term used to describe students using technology for nonacademic purposes (Gerow, Galluch, & Thatcher, 2010). Significance of Study The topic of technology usage and student demographics and learning is still in its infancy. With the smartphone barely a decade old, there are limited empirical and longitudinal studies focusing on the impacts of technology use on student learning, persistence, and success. The significance of this study addressed three important criteria: significance in the body of literature and research, significance in relation to theory, and significance in relation to practice. Research Researchers have studied academic achievement for decades specifically focusing on the demographic differences of college students in relation to student success and retention (Astin, 1964; Astin, 1997; Bayer, 1968; Braxton, 2000; Tinto, 1987; Tinto, 1998; Vaughan, 1968). The primary demographic metrics presented throughout the literature used to study academic achievement are race, sex, and grade point average (Reason, 2009). In addition, researchers have studied a wide array of other variables seeking to find predicting variables for student success. Some of these include age, economic class, academic preparation, and declared major to name a few (Keller, 2001; Murdock & Nazrul Hoque, 1999; Reason, 2009). Pascarella & Terenzini (1998) recommended that with the ever-changing demographics within higher education, researchers should continue to change how, who and why to research the student population. Furthermore, it is vital that, with the 10 increasing diversity in college students, researchers should attempt to understand how predicting variables interact with each other. Although the literature is relatively shallow on the topic of modern technology use related to academic performance, studies are starting to reveal that there may be some serious issues pertaining to technology use and the implications this use may have on academic performance. The outcomes of this study contribute to current literature presenting that modern technology use might be a wolf in sheep’s clothing; technology generally is accepted as a positive addition in educational settings, even though scholars are beginning to better understand the underbelly of modern technological inventions. These innovations create, for some, a tension to focus, higher levels of stress, and depression--which all can negatively impact a student’s ability to complete academic tasks and persist. The findings of this study are intended to offer insight into the relationship between IAT scores and student success as measured by grade point average. Findings also discussed the differences between student demographics of sex, research site, race, and student classification, and IAT scores. In addition to contributing to the understanding of links between demographics and IAT score, and IAT scores and academic performance, longstanding theories of student learning and engagement are also challenged. The following section continues this discussion. Theory Multitasking research and theories are not new. In fact, researchers have studied and theorized about multitasking for over half a century (Craik, 1948; Navon & Gopher, 1979; Telford, 1931; Welford, 1952). Additionally, researchers have studied the way 13 students seemed to constantly be on the phone...in class, at sporting events, in the cafeteria, at student events. All of this led the researcher, and fellow colleagues, to begin technology ‘fast’ weekends, encouraging students to take a break from technology for everything other than homework. While sitting at lunch with a mentor in 2006, she said something that is still haunting. She said, “It takes up to twenty years for some foods or drugs to get through testing and regulation before we can ingest it. However, Microsoft can roll out the newest version of Windows and push updates to our computers overnight, and we all ingest it without much thought, and with very little regulation or testing.” The question that has been mulling in the researcher’s mind for some time: What is all this modern technology doing to students? This has been the driving question throughout this doctoral program and is why the researcher chose this perspective for this dissertation. Anecdotally, the researcher has noticed changes in students over the years, but as a scholar/practitioner in training, the researcher began this doctoral program searching for literature and answers to the growing technology phenomenon. This study was a culmination of this quest. Chapter Summary With the rapid growth of technology innovation, the researcher was concerned that scholars are not truly able to keep up with the changes that are pumped out by technology giants and consumed by students. As the literature is beginning to show, technology use, while having plenty of positives, also has negative impacts. These negative impacts have arguably not received enough attention. This study intended to explore the student scores on the Internet Addiction Test and how they are related to 14 student demographics and academic achievement as measured by grade point average. The findings of this study are added to the body of literature and may provide valuable opportunities for practitioners to intervene. The next chapter, Chapter Two, provides a review of the literature examining technology in higher education and the implications of academic achievement. Chapter Two more thoroughly explores college-aged young adults (digital natives) with a discussion of demographics, an examination of the use of grade point average (GPA) to measure student achievement, multitasking literature, and both positive and negative impacts of technology on student learning and achievement. Furthermore, primary theories for this study are presented along with an analysis of inventories used to assess technology use. Chapter Two concludes with a discussion of Internet Addiction and Problematic Internet Use. 15 CHAPTER II REVIEW OF LITERATURE From 2000 to 2015, the Pew Research Center conducted a longitudinal study analyzing technology use in the United States. The findings presented rapid growth in the innovation, sales, and usage of technology (Perrin & Duggan, 2015). This technology surge was also evident across institutions of higher education. During this time, U.S. colleges and universities began implementing and utilizing technology throughout all areas of higher education (Christen, 2009; Costley, 2014). Whether laptops used in the classroom, online learning modules and courses enhancing distance learning, or online discussion groups and group projects, technological advances are impacting education. Additionally, both Educause and Pew Research reported that almost all college students accessed the Internet, used social network sites, and connected wirelessly with cell phones, tablets, and other personal computing devices (Hakoama & Hakoyama, 2011). In 2009, roughly ninety-five percent of college aged adults, categorized as ages 18-29, used the Internet. This was an increase from seventy-four percent at the turn of the 21st century (Derbyshire, et. al., 2013). Furthermore, in 2011, thirty-eight percent of students reported they could not go more than ten minutes without checking some form of technology, and seventy-three percent reported that they needed technology in order to study (Kessler, 2011). 18 researchers challenged future studies to continue to use gender as a predictor variable, however, to not focus on gender exclusively, but include other variables to get a better understanding and accuracy of results. Classification Numerous studies report that classification is a predictor variable of student success, even among traditional college students (18-25). This study uses the following classifications, 1st year students, 2nd year students, 3rd year students, and 4th year or more students. Students were asked to select which year in college they were when taking the survey. Classification is shown to be a significant variable in researching student persistence and achievement (Mayhew, et. al., 2016; Pascarella & Terenzini, 1998). Pertaining to technology use, students appear to use lower levels of non-academic technology as they matriculate through college (Junco, 2015). This study used classification as a demographic variable and explored the relationship between a student’s classification and IAT score. Sex and classification are two of the primary demographic variables used when studying college students. The next variable is race. Race Throughout the literature, race and ethnicity often are conflated (Reason, 2009). Although race is different from ethnicity in that race refers to a person’s physical characteristics compared to ethnicity which refers to cultural factors represented in a person; for clarity’s sake and alignment with the literature, race is used in this study. Research has found that race is a statistically significant predictor of student achievement and persistence (Peltier et. al., 1999, Reason, 2009). Whereas the issues are more 19 complex than just the race students are born into, African American, Hispanic, Native American and Pacific Islander students all tend to perform at lower levels academically and persist at lower rates when compared to Asian and White students (Murtaugh, et. al, 1999; Peltier, et. al., 1999; Reason, 2001; Reason, 2009). It is important to understand how students from different races are impacted by Internet addiction, thus the inclusion in this study. A fourth and final variable used in this study is grade point average. Grade Point Average (GPA) Grade point average, while criticized for its true representation of student capability or knowledge, is a metric generally accepted across higher education for gauging academic achievement. The literature shows that grade point average is a significant variable in studies analyzing student persistence and achievement. First-year GPA, along with high school GPA and cumulative college GPA all have been shown to predict college student achievement (Reason, 2001; Reason, 2009, St. John, et. al., 2001). Studies show that students with higher grade point averages at all levels of education have higher levels of academic success as compared to students with lower grade point averages. As with the previous three demographic variables, GPA should not be a standalone variable and researchers recommend that other variables be used with GPA to help find significant results (Reason, 2009). For this study, a student’s cumulative grade point average was analyzed along with a student’s IAT test score to explore the relationship between IAT score and GPA. Each of the demographic variables were used in this way to explore any relationships between them and a student’s score on the Internet Addiction Test. 20 For this study, students self-reported GPA. Although some question the validity of self-reported GPA, decades of literature provide support that self-reported GPA, particularly in college student populations, is as valid an indicator of future success as actual grade point average (Baird, 1976; Cole & Gonyea, 2010; Schwartz & Beaver, 2014; Sticca, et. al., 2017; Stone, et. al., 1999; Talento-Miller & Peyton, 2006). These primary variables, along with student classification and research site, comprised this study. The following section of this review of literature provides a better understanding and description of the contemporary college student. Many researchers are calling traditionally aged students today ‘digital natives’ and these students are discussed below. Digital Natives Contemporary, traditional college students have grown up in a world surrounded by numerous forms of technologies. Throughout their lives, computers have been accessible in the home, at school, or in libraries and, at a young age, the smartphone became a part of daily life. Many researchers call current 18-29 year old’s “Digital Natives” (Bowe, & Wohn, 2015; Dede, 2004; Prensky, 2001, Thompson, 2013). This age group is saturated with technology (Bowe, & Wohn, 2015; Rideout, Foehr, & Roberts, 2010; Thompson, 2013). Growing up with access to technology, and a level of competence with various types and forms of technology, may lead to an assumption that digital natives are able to handle the demands of technology and better understand it, as it relates to multitasking. Even though the broad characterization of technological fluency in this generation is generally accepted, scholars are finding the technological competency gap between 23 Multitasking can interfere with memory, which leads to lower academic performance (Hembrooke & Gay, 2003; Fried, 2008). Additionally, students who multitask report that they study less (Bowman, et. al., 2010, Fried, 2008). This lack of study time hinders academic success compared to students who study longer (Bowman, et. al., 2010, Fried, 2008). Also, digital natives who claim to multitask report higher levels of mental exhaustion as compared to peers who multitask at lower levels (Small & Vorgan, 2008). More Pictures, Less Text Students report higher levels of affinity in learning environments with more pictures and less text. This is especially true with online material (Tapscott, 2009). The desire for more pictures is creating higher levels of visual and spatial skills in digital natives (Prensky, 2001; Tapscott 2009). However, like a student’s desire for fast-paced environments, the desire for pictures also may limit deep, reflective reading and critical thinking skills (Carr, 2010). Nonlinear Information Processing Many scholars have studied the notion that learning can happen through many different platforms and through many different systems, as compared to previous decades. Technology has provided this learning enhancement and digital natives are now learning in nonlinear ways (Bowe, & Wohn, 2015; Dede, 2004; Tapscott, 2009). Using technology, students can find information quickly and from numerous sources. This simultaneous exploration may create greater understanding of complex ideas (Tapscott, 2009). A risk, however, is that when digital natives use nonlinear thinking they may 24 struggle when presented with tasks requiring linear thinking, such as accounting, forensics, and even scientific experiments (Carr, 2010). Collaborative Learning and Constant Connectivity College students are growing more collaborative and seek collaborative learning environments at greater rates (Rosen, 2010; Tapscott, 2009). Technological advances have largely fueled this desire for collaboration (Rosen, 2010). Students can work collaboratively outside of the brick and mortar education structures of the past. This collaboration inspires learning and teamwork and creates projects that might have been difficult to accomplish in previous generations (Prensky, 2001; Tapscott, 2009). A potential risk presented when using technology for connection is the distraction of available online socializing methods. Students report distractions, from social media for example, when studying and working online (Bauerlein, 2008). These distractions tend to limit academic success (Fried, 2008). Learning by Doing Rather Than Lecture or Reading Due to collaborative approaches educators create for college students, digital natives may experience gains in active learning. A benefit to this desire for active learning is that students are not waiting for instruction and are taking more responsibility for learning (Prenski, 2001). However, some students are becoming more apprehensive to traditional instructional approaches such as lectures or presentation, thus many important concepts and steps exposing essential content may be missed (Mayer, 2004). Furthermore, studies have shown digital natives struggle to learn in non-active settings (Tapscott, 2009). 25 Balancing Time between Work, Play and Leisure Digital natives appear to manage the demands of balancing schedules well (Ito, et. al., 2010). Students mix work and play and use time well to complete tasks. Studies show that mixing work and play can create more imaginative problem solving which ultimately enhances learning (Ito, et. al., 2010). However, at times, this desire to mix work and play creates an expectation for entertainment in the educational setting (Crede & Kuncel, 2008). This expectation may inhibit learning and create impatience in form academic settings (Crede & Kuncel, 2008; Mayer, 2004). Expectation for Technology in Educational Settings Contemporary students are demanding more technology integration in learning settings. This trend has been growing for two decades (Dede, 2004; Prensky, 2001; Tapscott, 2009). Much of this is rooted in the students-as-consumers literature. Digital natives have grown up in a buy-and-consume society (Hill, 2011). Therefore, students- as-consumers often expect colleges and universities to meet their needs and desires. This includes providing not only wireless networks, but also allowing the use of technological devices (Delucci & Korgen, 2002; Hill, 2011; Obermiller & Fleenor, 2005). Students who believe they are entitled to technology use do not feel remorse when cyber slacking. Consumers believe that they are paying for college and can use technology as they please (Taneja, Fiore, & Fischer, 2015). Educators who implement technology into pedagogy are showing positive results to student engagement and learning (Mishra & Koehler, 2009; Mishra & Koehler, 2010). However, with the implementation of technology, researchers are finding that higher levels of distraction are present with multiple streams of information present at one time 28 2009). These definitions encapsulate much of the psychological research analyzing the brain’s ability to focus on different tasks. To understand multitasking better, it is important to analyze the way human cognition works. The Adaptive Control of Thought-Rational model, created by J.R. Anderson (2007), describes human cognition as independent but interacting thought modules. These thought modules are called threads, and each thread can contain active tasks. Although the threads all run parallel with each other, Anderson (2007) posits that only one thread can be active or executing at any given time. Even though some studies show that multitasking may not be detrimental to routine or familiar tasks that require minimal cognitive effort (Just, Keller, & Cynkar, 2008), most of the research finds that multitasking is very difficult at best, if not impossible to do. The human brain is not adept at processing multiple streams of information at once (Marois & Ivanoff, 2005; Monsell, 2003; Ophir, Nass, & Wagner, 2009). The challenge for the brain, when it comes to multitasking, is that multitasking challenges both working memory and processing. Multitasking becomes detrimental to learning when the brain’s cognitive resources are depleted or limited due to the amount of information bombarding the brain (Kraushaar & Novak, 2010). Additionally, learning, especially more complex problems, requires a high level of cognitive processing (Mayer, & Moreno, 2003). The high level of processing required, alongside the brain struggling with multiple forms of information at once, all but eliminates the ability to multitask. Furthermore, when the amount of information and complexity of task exceeds the brain’s capacity, an individual’s ability to learn diminishes, along with lagging performance related to the task (Mayer, & Moreno, 2003). 29 In relation to this study, the research pertaining to multitasking is important because it helps shine light on the effects of students using technology while attempting to complete academic tasks. Previous studies have uncovered that students attempting to multitask experience distractions during lectures, and exhibit lower levels of success in the academic setting (Bellur, Nowak, & Hull, 2015; Cerretani, Iturrioz, & Garay, 2016; Junco, 2012; Zhang, 2015). These lower levels of success may be partly due to a lack of efficiency or a depth of learning. Students who multitask take more time to complete projects when compared to students who focus on a single task (Courage, et. al., 2015). Furthermore, students who are multitasking have a difficult time getting beyond superficial learning and into a deeper understanding of presented content (Courage, et. al., 2015). Understanding the literature on multitasking helps provide a foundation for the struggles students may find using technology in educational settings. Additionally, the literature is clear that technology has affected the way digital natives learn. What is still unclear, however, is to what degree these technological impacts are positively or negatively impacting learning. It is important, however, that researchers continue to study this generation of students and how the broad array of technological innovations are influencing digital natives. The following section better examines the positive and negative impacts of technology on today’s college students. Positive and Negative Impacts of Technology Use and Student Success The modern technology phenomenon is still in its infancy, with most of the explosion happening in the past ten to fifteen years largely due to the creation of the iPhone and social networking sites like Facebook (Christakis et. al., 2011). With this 30 growth in technology use, the literature discussing college student technology use and the impacts on student learning is divided and relatively shallow. Chen and Peng (2008) found that heavy Internet users, with more than thirty-four hours of online activity per week, had lower grades as compared to students who used the Internet less than thirty- four hours per week. This technology use is depicted as general use, which included both academic use and non-academic time spent on the Internet. Conversely, Keyser, Wentworth, & Middleton, (2014) conducted a literature review with mixed results on the negative effects of technology use on academic performance. The following sections outline studies and findings on both the positive and negative impacts technology use has on student learning and success. Technology and Positive Impacts on Student Learning Technology innovations affect higher education from course offerings, to the way business is conducted, to the way students learn. Technology has greatly influenced the presentation of knowledge and course material, the evaluation of activities and courses, the business of the university, and the ways in which research is conducted (Engstrom, 1997). It is evident that educators, administrators, and students use technology across higher education. How education is offered or provided to students significantly changed with the introduction of technology to higher education (Adams et. al., 1999). Historically, in order to go to college, students had to attend brick and mortar classrooms. Currently, technology is bringing college students and higher education together, with colleges offering increasing numbers of programs online and through distance learning (Kenney, 2011; Kurt, 2010). 33 Gaffney, 2008). Conversely, students using laptops in class also experience more distraction and perform at lower levels of academic success compared to students not using laptops (Fried, 2008). Students describe using laptops to surf the web, watch movies, play games and check social media while in class (Lauricella & Kay, 2010). Interestingly, using laptops in class affected more people than just the student with the laptop. Sana, Weston, & Cepeda (2013) discovered that students sitting near someone using a laptop performed worse on tests as compared to students without the distraction of the laptop. Studies attribute this lack of success to the distractions created by the technology (Hembrooke & Gay, 2003). Laptops are one of the oldest forms of modern technology used by college students, however, they are not the only distraction contemporary students face. Researchers are finding that a newer form of technology is also directly competing for a college student’s attention. Mobile phones have become prolific throughout higher education and the following section examines how these personal technology devices are impacting students. Mobile phones. Radio and television are examples of intellectual technologies, or technologies that stretch brain functions. Mobile phones are an example of networked technologies. Scholars classify network technologies as an extension of intellectual technologies (Misra, et. al., 2016). Interestingly, studies show networked technologies absorb other intellectual technologies (Carr, 2010). In this way, mobile phones have affected society arguably as much as any previous form of technological innovation. Mobile phones provide vast arrays of information in the palm of the hand. Phones have taken the place of maps, watches, and television, just to name a few of the popular functions. These functions have led scholars to posit that mobile technologies 34 create an absent presence (Gergen, 2002; Stone, 2007). The notion of absent presence is that a person is physically in one place, but because of technology, they are mentally elsewhere. This absent presence creates situations where mobile phone users are occupying two realities at once: a virtual reality and a present reality (Misra, et. al., 2016). Due to these dual realities and the tension to function between them, studies are showing that two primary implications are present: microsocial fragmentation (Gergen, 2003) and horizontal relationships (Gergen, 2002). Microsocial fragmentation. Mobile phones allow the user to manage multiple social groups such as family, friends, and work colleagues, at one time. Proximity and communication challenges of the past are virtually eliminated, and users can connect with people in real time all around the globe. Users think about this unrestricted connection whether using the phone or not (Srivastava, 2005). These connections create tension in the brain and frequently subsume other brain functions, making it difficult for the individual to be present. One study showed that people in small groups checked their mobile devices every three to five minutes, even if it did not buzz or ring (Misra & Genevie, 2013). Along these same lines, numerous studies show an emerging phenomenon called “phantom vibration” where the users think their phone is vibrating in a pocket or bag, when it actually is not (Drouin, Kaiser, & Miller, 2012; Lin, Lin, Li, Huang, & Chen, 2013). The phantom vibrations and constant impulses people feel to check phones creates a cognitive tension that challenges the present reality. This tension creates distraction and withdrawal from real-time, present relationships; this is frustrating to friends and acquaintances in proximity (Humphreys, 2005). This frustration and tension 35 oftentimes fracture in-person interactions and relationships because distance relationships, conversations, and other concerns are salient (Turkle, 2012). For college students, this microsocial fragmentation is challenging. As mentioned previously, students may become distracted in class by phones even if they do not buzz or ring. Furthermore, the virtual connections provided by burgeoning technologies may strain social relationships in class, on campus, or at events. Students may feel connected to many people, but long for the deeper, real interpersonal relationships (Gergen, 2002). This is an example of how horizontal relationships challenge vertical relationships. The next section discusses this shift from vertical relationship to horizontal relationship. Horizontal relationships. The divided attention technology has ushered into society has created a societal shift from vertical relationships to more horizontal relationships (Misra, et. al, 2016). Superficial, shallow commitments that take relatively little to no effort or attention depict horizontal relationships. Conversely, vertical relationships are deeper and more meaningful. In order to strengthen vertical relationships time is required, along with commitment and many times some sacrifice (Gergen, 2002). Conversations in horizontal relationships are brief, simple and rarely require follow up. Some call these horizontal conversations sound bytes. Mobile phones and other technologies continue to encourage these types of conversations and relationships (Gergen, 2002; Turkle, 2012). This societal shift towards horizontal relationships has strained basic tenets of compassion, empathy and deeper understanding in culture today (Immordino-Yang, et. al., 2009). The type of introspection and processing necessary for empathy and compassion to occur is typically slower and more involved. Oftentimes, vertical 38 College students distracted by or utilizing non-academic technology at high levels are more likely than their peers to fall behind in school, which leads to lower levels of persistence (Armstrong, Phillips & Saling, 2000). Specifically, students spending higher amounts of time on the Internet or mobile phones have a lower self-confidence and score lower on emotional intelligence inventories when compared to peers who spend less time on the Internet and mobile phones (Beranuy, et. al., 2009). These same students also have lower retention rates when compared to peers using less non-academic technology (Beranuy, et. al., 2009). These lower retention rates may lead some students to suffer psychological issues. The next section will review some of the psychosocial issues students face. Technology use and psychosocial issues. In addition to the strain technology use has on academic success, numerous studies are focusing on troubling psychosocial byproducts of extended technology use among college students. Moderate to severe levels of Internet Addiction may lead to a range of psychosocial issues in college-aged young adults (Derbyshire, et. al., 2013). For example, college students using technology more than their peers exhibit higher levels of stress (Dick, 2013; Kim, et. al., 2007; Pennebaker, et.al., 2001; Turner, et. al., 1995). Many students today are plugged into various forms of technology, and the stress associated with technology use and in more severe cases, technology addiction, is creating negative experiences such as loneliness and depression (Turel, 2015; Velezmoro, Lacefield, & Roberti, 2010; Wei, 2001). Research is also discovering connections between higher levels of screen time associated with lower levels of psychological well-being. Psychological well-being is comprised of happiness, self-esteem and overall satisfaction measures (Twenge, Martin, 39 & Campbell, 2018). Students spending more time on technology and less time on non- technology activities such as sports or clubs, social interactions, and religious activity reported lower levels of psychological well-being (Twenge, Martin, & Campbell, 2018). Researchers are finding links between stress, psychological well-being, and Internet addiction. As a student’s Internet addiction increases, depression and stress scores increase significantly (Derbyshire, et. al., 2013). This relationship is troubling because students presenting symptoms of higher levels of stress are less likely to persist compared to students with lower stress levels (Krumrei-Mancuso, Newton, Kim, & Wilcox, 2013; Velezmoro, Lacefield, & Roberti, 2010). One body of literature linking technology to higher levels of stress and depression is the study of social media use and is discussed further in the following section. Social media impacts. In 2011, over ninety percent of college students used Facebook, with fifty-eight percent using it multiple times a day (Dahlstrom, 2011). Extending the study to include other forms of social media (Twitter, Instagram, etc.), that number increased to nearly ninety-five percent (Dahlstrom, 2011). It is safe to say that college students, along with large portions of society, are using social media. Interestingly, as students matriculate through college, it appears they may utilize social media at lower amounts. In 2015, college seniors were shown to spend less time on social media as compared to first-year students. In fact, as students moved from first- year students to second to third, each year was associated with lower levels of social media use (Junco, 2015). Students often recognize their personal technology use is interfering with academics. Students who used Facebook regularly admitted that they studied less than 40 peers who they perceived used Facebook less than them (Wentworth & Middleton, 2014). Seventy-nine percent of these students, however, felt that their social media use was not affecting academic performance (Wentworth & Middleton, 2014). This same study found a negative correlation between the time a student spent on social media and grades. The authors found that a student’s self-report pertaining to technology use had little to no impact and was not accurate to the findings (Wentworth & Middleton, 2014). There are conflicting bodies of literature discussing social media use and education. Between 2008 and 2010, researchers reported neutral or even positive effects of technology use in relation to academics. In 2009, a study found that Facebook use had no statistically significant impact on a student’s grades (Pasek, More & Hargittai, 2009). Hargittai and Hsieh (2010) found similar results in that students using various forms of social media showed no difference in academic performance when compared to peers not using social media. Likewise, a study presented in the NASPA journal in 2008 presented a positive relationship between a student’s grade and moderate amounts of Facebook use (Kolek, & Saunders, 2008). The literature documents the negative impacts that technology use has on students quite well. The following section seeks to better explore how these negative impacts of technology use impact student success. Using theory to help illuminate these impacts is a critical step in a research project. The following section outlines and discusses the theories that influenced this study. Relevant Theories Two theories influenced the design of this study exploring the relationship between a student’s addiction to technology and the Internet and academic performance 43 Distraction-Conflict Theory Zajonc’s Distraction-Conflict (DC) theory is a tenet of the broader Social Facilitation theory (Sanders, 1981). Distraction Conflict analyzes how individuals work with distractions. This theory surmises that when distractions are present, they create a physiological arousal in a person. Initially, distractions help an individual focus and perform better on easy, rudimentary tasks; however, as the tasks become more complex, the individual begins to struggle managing the distraction and the task at hand (Sanders, 1981). Distractions can be any stimuli, social or non-social that does not align with the task. The stimuli can be external or internal to the individual, created by the individual or another party. These distractions create an attentional conflict where a person must decide how to focus attention (Sanders, 1978). Zajonc (1965) believed that people respond to situations largely in one of two ways. Dominant responses are responses that are used most often, thus the term dominates the hierarchy of responses. These dominant responses are oftentimes second nature, and because of the number of times an individual uses these responses, they are easily reproduced. Zajonc further described a second type of response. These responses are used, but much more sparingly by the individual. These responses are coined non- dominant responses. Zajonc concludes that when distractions happen, the physiological arousal increases the tendency to use a dominant response. The more complex the task, or the greater the distraction, the less likely that a non-dominant response is used (Zajonc, 1965). To understand Distraction-Conflict Theory better, it is important to review three key components of the theory. These three components are social facilitation, cognitive 44 load and working memory, and task complexity (Nicholson, Parboteeah, Nicholson, & Valacich, 2005). Each of the three plays an important role in Distraction-Conflict Theory and is presented more in depth in the following section. Social facilitation. Social facilitation centers on how a person performs when encountered with another person. When a person focuses on a task and another person enters, this creates a physiological arousal and a distraction. This is also known as social facilitation. For some, these social distractions help complete the task, while for others, these social distractions can hinder performance (Nicholson, Parboteeah, Nicholson, & Valacich, 2005; Sanders, 1978). The social distractions create a cognitive load which impacts working memory and performance. Depending on the complexity of the task, a person might struggle to facilitate social distractions while attempting to complete the task. In this way, social facilitation closely relates to cognitive load, which is discussed in the next section. Cognitive load and working memory. Cognitive psychologists describe cognitive load as a level of mental activity at any given time that imposes on the working memory of a person (Nicholson, Parboteeah, Nicholson, & Valacich, 2005; Sweller, 1994). Working memory, also known as short-term memory, is directly related to how quickly one processes information or thinks about things. Working memory is the temporary storage files of the brain. Cognitive load directly affects working memory. When too many distractions bombard a person, the cognitive load increases and taxes the working memory. When this happens, a person is oftentimes challenged with choosing a focus, which in the presence of more complex tasks, limits the ability to complete the task (Nicholson, Parboteeah, Nicholson, & Valacich, 2005). 45 Task complexity. The final component of the Distraction-Conflict Theory is task complexity. Task complexity is easily defined as the level of complexity or difficulty of the task at hand. This component is relative to the individual and is based on several factors including aptitude, experience with the task, and other internal and external factors present when the task is presented (Sweller, Van Merrienboer, & Paas, 1998). Task complexity relates to cognitive load in that the easier the task, the lower the cognitive load. Conversely, the more difficult the task, the greater the cognitive load. Better understanding the three components of social facilitation, cognitive load and working memory, and task complexity helps one understand that a person’s performance will vary greatly depending on these factors. In general, the more complex the task, the more a person will struggle when confronted with distraction (Nicholson, Parboteeah, Nicholson, & Valacich, 2005). In this study the assumption is that college lectures and academic work (reading, writing, and research) are complex tasks for many, thus distractions may cause a person to struggle. As previously discussed, non-academic technology use creates a distraction. Studies are also indicating that technological distractions might challenge cognitive load and working memory more than other distractions, increasing the difficulty of highly complex tasks when confronted with non-task technology (Fockert, 2013). Because the human brain cannot multitask, these distractions will likely negatively affect performance. Rational Addiction and Distraction-Conflict help describe struggles college students arguably face when presented with technology that competes for attention. In addition to the distraction technology presents to college students, recently higher 48 twenty-question inventory (Widyanto, Griffiths & Brunsden, 2011). The initial studies using the IAT began to receive national publicity with results published in The Wall Street Journal, The New York Times, and the London Times (Young, 1998). The IAT uses a six-point Likert scale and divides respondents into four categories. Respondents scoring between 0-30 points reflect normal Internet usage. Scores of 31-49 represent mild Internet addiction. Respondents scoring 50-79 represent moderate levels of Internet addiction and scoring 80-100 represents a severe dependence on the Internet (Young, 1998). Appendix A contains the questions for the IAT along with the scoring rubric and instructions. Numerous studies tested the IAT and found the instrument valid and reliable. (McMurran & Widyanto, 2004; Widyanto, Griffiths & Brunsden, 2011). A factor analysis revealed strong internal consistency and concurrent validity with six factors: salience, neglecting work, neglecting social life, excessive use, lack of control, and anticipation (Widyanto, Griffiths & Brunsden, 2011). The most reliable of these six factors was salience. The IAT was one of the first assessments created to assess Internet addiction. Because of this, it is one of the most popular and is still used by researchers. The IAT, however, is not the only instrument available. The following section will provide a cursory overview of five additional instruments created to analyze the Internet addiction phenomena. These were chosen to help provide perspective on the instruments available for this study. 49 Pathological Internet Use Scale The Pathological Internet Use Scale was created by researchers Morahan-Martin and Schumacher (2000) to conduct research on Internet addiction. This study created the term “Pathological Internet Use” (PIU) instead of Internet addiction (Morahan-Martin & Schumacher, 2000). This specific study used thirteen questions and focused more on the behaviors of PIU. Coupled with the UCLA Loneliness scale, the study found that a little over eight percent of respondents experienced PIU. The Pathological Internet Use Scale has high levels of internal validity (Morahan-Martin & Schumacher, 2000) and posits that users scoring higher levels of PIU chose to have social interactions online instead of in person. Additionally, higher PIUs felt much more competent and comfortable in online social settings as compared to face-to-face interactions. Generalized Problematic Internet Use Scale (GPIUS) Using the Davis’ (2001) Problematic Internet Use theory, the Generalized Problematic Internet Use Scale was created to conduct a study on undergraduate students at the University of Delaware (Caplan, 2002). Like the Pathological Internet Use Scale, Caplan (2002) administered the GPIUS with the UCLA Loneliness scale, along with three other psychometric measures. Three hundred and eighty six University of Delaware students took part in the research project. Only one finding was reliable and that was that shy students tended to use the Internet for socializing more than face-to-face contact. Internet Addiction Scale (IAS) The Internet Addiction Scale was developed and initially distributed in Canada in 2004 (Nichols & Nicki, 2004). Like the IAT, the IAS used the substance dependence 50 criteria of DSM-IV in the creation of the 31-question inventory (American Psychiatric Association, 1994). Unlike the IAT, which had six factors prove reliable, the IAS only had one reliable factor, salience (Nichols & Nicki, 2004). Another concern was that compared to the other assessment inventories, the IAS only found one percent of respondents dependent to the Internet as compared to a thirteen percent average of other inventories (Nichols & Nicki, 2004). Internet Addiction Tendency Scale In 2004, researchers Song, Larose, Eastin and Lin studied the difference between process gratification and content gratification as it relates to the tendency to become addicted to the Internet (Song, Larose, Eastin, & Lin, 2004). Content gratification focused on an individual's pleasure from the material on the Internet, while process gratification focused on merely the practice of using the Internet. The researchers conducted the study at both the University of Michigan and Ohio University with 498 combined respondents. Researchers created the Internet Addiction Tendency Scale to conduct the study. Through factor analysis, the researchers found only one factor to be significant: Information Seeking. Furthermore, the factor of diversion, which the authors predicted as unrelated, was found to be significant (Song, Larose, Eastin, & Lin, 2004). Internet Effect Scale (IES) Two researchers created the Internet Effect Scale to conduct a study in Pakistan. The IES is included in this review of assessments for two reasons. The first is it is a more recent attempt at assessment as compared to the others and, secondly, it attempted to analyze both the positive and negative effects of Internet use (Suhail & Bargees, 2006). 53 decision makers in higher education must be aware of these factors and make decisions accordingly to help institutions move forward in a world entrenched in technology and its use. It is becoming evident that technology is significantly affecting college students today. The question that is arguably more pertinent than ever is how is technology affecting students? This study provided an examination of how technology use and in some cases Internet addiction impacts college students using the IAT to gauge Internet use and addiction, and explored relationships between the IAT score, academic achievement, and student demographics. This study may offer initial insights to higher education decision makers, and provide a better understanding of the implications of technology use among contemporary college students. This understanding may help identify students who are addicted to the Internet and potentially negatively impacted by individual technology use. 54 CHAPTER III METHODOLOGY Research Context Individual technology use is increasing rapidly among college students with students reporting that not only they use technology at increasing levels, but they are becoming more reliant on forms of electronic technology to study and stay connected with peers (Derbyshire, et. al., 2013; Kessler, 2011). With technology prevalent throughout higher education, researchers are questioning how the increased technology exposure and use is impacting college students. This chapter discusses the research study’s design, which includes the research perspective, purpose statement, research questions and hypothesis. Research population and intended sampling are also discussed, followed by presentation of the study’s methodology and instrument. Finally, data collection procedures, data analysis, limitations, and delimitations of the study are provided. Research Design Research Perspective Crotty (1998) defines epistemology as “how we know what we know” (p.8). Meaning existing independently from human conscience is a primary tenant of objectivism (Crotty, 1998). Objectivists actively analyze and look for facts when seeking truth (Crotty, 1998). Although seeking objective truth, post-positivism suggests that 55 knowledge depends on human interpretation and experiences, making it difficult to find an absolute truth (Creswell, 2014). Post-positivism is centered on seeking understanding for regular, observed phenomena (Crotty, 1998). Based on this, the researcher positioned this study using an objectivist epistemology with a post-positive theoretical perspective. Along with the objectivism and post-positive alignment, this study used a quantitative design and explored the relationship between the Internet Addiction Test scores and student success as measured by grade point average. Furthermore, the study analyzed if the IAT score was predictive of grade point average, and sought to find if there were differences between reported IAT scores and student demographics including sex, research site, race and student classification. Purpose Statement The purpose of this quantitative study was to analyze the relationship of the Internet Addiction Test (IAT) score and academic performance and identify difference between student demographic variables and IAT scores. This study used Young’s (1998) Internet Addiction Test (IAT). The IAT score was used to determine a student’s addiction to the Internet. A student’s academic performance was measured by grade point average. A student’s demographics included sex, research site, race, and student classification. Research Questions and Hypothesis Informed by rational addiction theory and distraction conflict theory discussed in Chapter Two, this study tested the relationship between a student’s IAT score and grade point average, and analyzed if there was an influence between traditional college students’ IAT scores and grade point averages was analyzed. Additionally, the study examined if there were any differences between the demographic variables of sex, 58 Ha/Alternate hypothesis: There is a significant difference of mean of IAT score and student classification. Based on the IAT literature previously discussed in Chapter Two, it was anticipated that the null hypothesis for this question would be rejected and there would be a significant difference in the means of IAT and student classification. The literature presents that upper division students tend to use the internet for nonacademic use less than lower division students. Research Population, Sampling, and Data Collection Population Traditional aged college students commonly categorized as students between the ages of 18-25 years of age are a subgroup of students attending higher education institutions. As discussed in Chapters One and Two, this same age group, called digital natives by some scholars, has grown up in a world surrounded by different forms of technology. For this reason, the population of the study was traditionally aged (ages 18- 25) college students attending college in Oklahoma during the spring 2019 semester. Traditional aged students from both public and private institutions in Oklahoma participated in this study. The institutions selected for this study were a public, four-year research institution with more than 20,000 students, a private, four-year, liberal arts institution with more than 3,000 students, and a religiously affiliated, private, four-year liberal arts institution with more than 2,000 students. These schools were selected to create a broad sample from the region of traditional college students from both public and private institutions in a regional setting. One of the variables in the study is research site, thus the selection of both public and private institutions. 59 Sampling As previously outlined, this study used convenience sampling within the population requirements. Convenience sampling is a non-probability sampling method (Gay, et. al., 2012). Because the number of all traditional, public and private college students across the United States is large and may differ regionally, the researcher decided to focus the study on a regional selection of schools, making convenience sampling the method selected for the study. Convenience sampled participants were participants who were available at the time of the study, were willing to participate in the study, were accessible, and who met the criteria of the study in relation to the population parameters (Gay, et. al., 2012). The researcher worked with the Institutional Research Board at each site to collect email lists of students 18-25 years of age. Emails were crafted for each participating institution and sent to each institution separately. However, the text of the emails was identical regardless of institution. All convenience sampled participants received an email with a link to the Qualtrics survey in the spring of 2019. Students were informed that the study was voluntary and were asked to electronically sign an IRB-approved informed consent form prior to completing the questionnaire by clicking on a radio button at the bottom of the consent form (Appendix B). The questionnaire (IAT) included twenty questions focusing on an individual’s Internet usage. Students self-reported their cumulative grade point average. A discussion on self-reporting GPA is included in Chapter Two. In addition, biographical information questions including sex, research site, race, and student classification were included in 60 the instrument. Finally, there were three internal validity questions. The instrument is included in its entirety in Appendix A. Methodology and Instrument Methodology This study utilized a quasi-experimental, cross-sectional, quantitative design. A cross-sectional design analyzes data from a population or representative sample of a population at a given point in time (Creswell, 2014). Data were collected through a web- based survey that was disseminated to student email addresses. The data were collected and stored in a password protected account. Only the primary researcher, his advisor, and the OSU IRB (if requested) have access to the data. Young’s (2008) IAT scoring metric allotted a score of 0-100 to each respondent that placed them in a range from normal Internet use to severe dependence on the Internet (100 total points). SPSS version 24 was utilized to examine, first, if there was a relationship between the student IAT score and GPA, second to study if the IAT was predictive of GPA, and finally, compared differences of the demographic means and the IAT scores. A Pearson correlation determined if a student’s IAT score and grade point average were related, an OLS test (linear regression) determined if the score on the IAT predicted grade point average, and an ANOVA identified differences of means between the biographical variables and a student’s IAT score. Instrument This study used Young’s Internet Addiction Test (IAT); the IAT was created by Dr. Kimberly Young at St. Bonaventure University in 1998 (Young, 1998). Dr. Young created the IAT to assist with the intake of clients seeking support from the Center for 63 The three factors in the study were: Factor One, measuring psychological and emotional conflict and it accounted for 42.7% of the variance. Factor One also produced the highest Chronbach’s alpha (α = 0.93), which assumed high reliability. Factor Two measured time management conflicts and accounted for 8% of the variance, also with a high Cronbach’s alpha score (α = 0.86). Factor Three measured salience in terms of mood modification and accounted for 5.6% of the variance. Like the other two factors, high reliability is assumed with a high Cronbach's’ alpha score (α = 0.86). Additionally, each of these three factors showed strong internal consistency (Wiyanto, Griffiths & Brunsden, 2011). The Wiyanto, Griffiths and Brunsden (2011) study also ran correlations between variables with age and frequency of Internet use showing significant correlations. Age is significantly correlated to time management issues (time-management issues, r= 0.18; p < 0.01), and frequency of Internet use is significant to time management issues and salience (time-management issues, r= 0.26, p < 0.01; salience in terms of mood modification, r= 0.18; p < 0.05) (Wiyanto, Griffiths & Brunsden, 2011, pg. 145-146). Overall, the Wiyanto, Griffiths and Brunsden (2011) psychometric study found that time spent on the Internet was positively correlated with the IAT score, which suggests that the more time a user spends on the Internet, the higher likelihood of Internet addiction. The study also found that males tended to score much higher than females on the IAT and experienced higher levels of Internet addiction (Wiyanto, Griffiths & Brunsden, 2011). In addition to this study, the following studies produced similar results focusing on validity and reliability. 64 Widyanto & McMurran (2004) explored the psychometric properties of the IAT. This study used factor analysis to study the six primary factors of the IAT. The six factors examined were lack of control, excessive Internet use, salience, putting off work, neglecting social life and lack of self-control. The study found good internal consistency and validity. The study used a Pearson’s correlation and found that all six factors were significantly correlated with ranges from r = 0.62 to r = 0.226, p<.05, in the two tailed test. The two strongest factors in the study were excessive use (Cronbach’s α = 0.77) and salience (Cronbach’s α o= 0.82). The study concluded that the IAT was a reliable instrument and could be used for studying Internet addiction. Pawlikowski, Altstötter-Gleich, & Brand (2013) studied the validation and psychometric properties of a short version of Young’s Internet Addiction Test. The study addressed the factorial structure of the IAT. Using factor analysis to assess the IAT, the study found that the IAT has sound psychometric properties and the key elements and factors are valid and reliable with a Cronbach’s α of .897. Based on this study, the researchers believe that the IAT is useful for gauging Internet Addiction. Jelenchick, Becker, and Moreno (2012) assessed the psychometric properties of the Internet Addiction Test (IAT) as it relates to U.S. college students. This study used exploratory factor analysis to study the IAT along with 215 college students. Eighty- eight percent of the respondents tested as “average Internet users.” Twelve percent tested as “problematic Internet users.” The study found significance with Internet addiction in two factors: dependent use and excessive use. Dependent use was classified as social withdraw or awkwardness due to a preoccupation with the Internet and a Cronbach’s α = 0.91. Factor two, excessive use, was classified as loss of control or overuse of the 65 Internet with a Cronbach’s α = 0.83. Additionally, the researchers claimed that the IAT instrument is valid, reliable and could be used to study Internet addiction in U.S. college students. Frangos, Frangos, & Sotiropoulos (2012) conducted a study on the reliability of Young's Internet Addiction Test. This study was a meta-analysis of twenty studies with over 6,800 respondents using the IAT. Using the Cronbach’s values in each study, the researchers found that the overall Cronbach’s α = .889 and that the IAT is a valid and reliable instrument. Each of these studies has provided support for validity and reliability of Young’s Internet Addiction Test. Currently, the IAT is one of the most commonly used instruments for researchers studying Internet addiction. The IAT, however, is not immune from critique. The following section explores the critiques of the IAT. IAT Critique. The IAT has received criticism in two main areas. The first area of criticism is that of self-reporting. Beard and Wolf (2001) questioned validity of the instrument when so many of the questions are based on the assumed objectivity of the respondent through self-reporting. Additionally, Beard and Wolf (2001) expressed concerns regarding the use of Pathological Gambling criteria used to model the first eight questions. A question was posed as to whether these criteria were the best choice for gauging Internet addiction. Although these critiques raise concerns, the studies outlined previously (Frangos, Frangos, & Sotiropoulos, 2012; Wiyanto, Griffiths & Brunsden, 2011) suggest that the method of self-reporting does not influence the validity of the results. These studies demonstrate the IAT is a significant and reliable instrument for gauging Internet 68 Data Collection Upon approval by the IRBs at all institutions, an email with the electronic Internet Addiction Test (IAT) form and supplemental questions was sent to each student in the population via an institution email address. When students clicked on the link and prior to entering the survey and taking the IAT, students were presented with a cover letter that contained the consent statement and an overview of the study’s purpose statement. Students clicked on an “I Consent” icon which served as an electronic signature to participate in the study. Students who did not wish to complete the survey and clicked the “I do not consent” link were directed to a thank you page and did not complete the survey. Appendix B has a copy of the cover letter and consent statement. After agreeing to the statement of consent, students read the instructions for the IAT. The instructions are as follows: “The questionnaire consists of 20 statements. After reading each statement carefully, based upon the 6-point Likert scale, please select the response 0, 1, 2, 3, 4 or 5 which best describes you. If two choices seem to apply equally well, circle the choice that best represents how you are most of the time during the past month. Be sure to read all the statements carefully before making your choice. The statements refer to offline situations or actions unless otherwise specified. In addition, you will be asked basic demographic questions and a question regarding the technology you use in class.” The survey should have taken each respondent five to ten minutes to complete. Two follow up emails were sent to the students who did not respond within two week 69 increments (a total of three emails). No incentives were provided for this study. Respondent data were confidential and limited personally identifying information was collected by the researcher. Only the researcher, his advisor and the OSU IRB (upon request) have access to the data. The surveys were completed online using Qualtrics software and asked for limited identifying information including basic demographics of self-reported grade point average, sex, race, year in school (classification), and research site. After the data were collected the variables were recoded in the working data set as outlined previously. The code book in Appendix D contains the codes for this study. Survey data were kept in the password-protected account of the primary investigator (PI), and the data to be used in subsequent analyses were downloaded only to the password-protected computer of the PI. Only the PI, advisor and IRB has access to the completed survey data. The principal risks in this study are those associated with a breach of confidentiality concerning the respondent’s involvement in the research. Data Analysis The following quantitative statistical measures were used to analyze the data. First, the data collected were analyzed to ensure that there was not any corrupted or incomplete data sets. This was done by importing the data collected in Qualtrics into IBM’s Statistical Package for Social Sciences program version 24 (SPSS). The completed surveys were sorted in the database and all incomplete data sets were removed. Furthermore, any dataset not answering at least two of the three internal validity check questions correctly was discarded. 70 Once the data set was validated and complete, all data analysis was conducted using IBM’s SPSS version 24. A descriptive analysis, Pearson’s r analysis, linear regression analysis, and analysis of variance was conducted for this study. The following sections describe this process of analysis. Descriptive. The descriptive analysis provides an overview of the sample of the population that completed the study. Descriptive statistics analyzed sex, research site, race, student classification, grade point average, IAT score. The descriptive statistics include number of participants, percentages, and means and were included in tables and graphs to provide a quick visual representation of the data sampled. Correlation. Next, a Pearson correlation was used to measure the strength of relationship, if any, between the variables. Pearson r provided an estimate for both the direction and the strength of a linear relationship (Gay, et. al., 2012). The Pearson r range is +1.0 to -1.0. A result of 0.00 results from two variables that are independent from each other, or that do not have a linear relationship. If the result is a negative integer, then the relationship is negative. Conversely, a positive integer is the result of a positive relationship (Lomax & Hahs-Vaughn, 2012). For example, in this study one hypothesis was that the higher a student scores on the IAT the lower the grade point average. In this case the Pearson r score would be a negative integer, showing a negative relationship between IAT score and grade point average. Regression. The next step of analysis for this study used an OLS analysis (linear regression) to test whether a student’s IAT score significantly influenced grade point average. The predictor variable IAT score was tested with the dependent variable of grade point average. The purpose of a linear regression was to test the influence of the 73 Classification – Underclass (1st and 2nd year), Upperclass (3rd, 4th, 5th year); Ethnicity – White, Non-White; Research site – Public, Private. Limitations and Delimitations Research, in its very nature, is conducted with a specific focus and with specific variables and focused parameters. This section discusses and accounts for the limitations and delimitations in this study. Limitations This study, like other research studies, has limitations. One such limitation was the type of technology a student had and used. Students have a variety of technology devices at their disposal and the number of devices likely varied among respondents with some having and using many devices, whereas others may only have and use one or two. While the number of devices may or may not have an impact on student success, this study was more concerned with the IAT score, and the literature is not clear on any relationships between number of devices and IAT score. Another limitation of this study was socioeconomic status. Socioeconomic status may inhibit a student’s ability to have technology. A larger sample size and choosing a more regional population was an attempt to control for socioeconomic differences. An additional limitation acknowledged in this study was how competent a student was with individual technologies. Some students are highly competent with various forms of personal technologies while others are not. For this study, the level of competence was not studied, rather the links between IAT score and grade point average, along with other student demographics and IAT score was researched. 74 Another limitation was a smaller sample size. Even with collecting data from three institutions, the N was likely to be under 1,000, making it more difficult to generalize the findings within the three institutions. However, it was believed that this study will begin to pave the way for future studies seeking to find relationships between technology use and student success. Because the body of literature is still shallow on the topic of college students and personal technology use and any influences on student success, this study will hopefully help lay some groundwork for future research. Since research on Internet addiction and college student achievement is limited, finding an instrument for this study was challenging. As discussed previously, the IAT was selected based on the literature showing it as one of the most utilized and long standing instruments utilized. However, Young’s (1998) intent when creating the IAT was as an intake form for counseling clients. This population differs from traditionally aged college students and the IAT may not be a ‘perfect fit’ to analyze students’ Internet addiction and use. Furthermore, technology has changed drastically since the inception of the IAT. The proliferation of social media, smartphones, and apps make it difficult to analyze what specifically distracts or challenges college students’ attention, and the IAT strongly relies on time spent on the internet to assess addiction – a relationship which may not be as simple as suggested by the design as the IAT. To date the IAT is as good an instrument as is available, as discussed previously in Chapter Two, but numerous limitations to use of the instrument are acknowledged. A final limitation is the assumption of autocorrelation was not met prior to analyzing the data using a linear regression. The autocorrelation assumption violation occurs when the residuals appear to be autocorrelated with each other or may be 75 influenced with other residuals (Lomax & Hahs-Vaughn, 2012; Nolan & Heinzen, 2012). Failure to meet the autocorrelation assumption is a liability for this study because the significance found in the regression may not be as strong or the results may not be significant at all due to the possibility of autocorrelation (Lomax & Hahs-Vaughn, 2012; Nolan & Heinzen, 2012). While the other three assumptions for regression were met, the findings of the regression portion of the analysis should be interpreted with caution, noting this failed assumption. Delimitations As mentioned, every study has parameters and that was the case for this study. This section outlines the choices made by the researcher regarding the boundaries for the study. One delimitation for this study was that the sample group chosen was traditionally aged college students. This sample excluded some college students, more specifically, those older than 25 years of age and under 18 years of age. This was intentionally done to help focus the research on a generation of traditional students at similar developmental stages experiencing common technological advances while in higher education. As previously discussed, some call these students digital natives. Studying students only in the Oklahoma geographic region was also a delimitation. As was discussed in the limitations section, researching a regional population was an attempt to account for broad socioeconomic differences that might present using a national population. To account for this delimitation, the researcher chose more than one institution along with different types and sizes of institutions in Oklahoma. This was done to achieve a larger, more diverse population of private and public, traditionally aged college students in the region. 78 CHAPTER IV ANALYSIS OF DATA Chapter Overview Chapter Four continues the discussion introduced in Chapter Three pertaining to the methodology and data collection of the study. The purpose of this quantitative study was to first analyze the relationship of the Internet Addiction Test (IAT) and academic performance, then study what influence the IAT score has on academic performance, and finally analyze if there is a difference between student demographic variables and IAT scores. This study used Young’s (1998) Internet Addiction Test (IAT). The IAT measured the level of a student’s addiction to the Internet. A student’s academic performance was measured by grade point average. A student’s demographics included sex, research site, race and student classification. This chapter is divided into three primary sections. First, a review of the research questions is presented along with the accompanying hypotheses. Additionally, in this first section, the specific statistical analysis utilized for each question is offered. Next, a description and overview of the respondents in the study are presented. Finally, the results from the data analyses are presented. Chapter Four is a bridge to the final chapter containing more robust discussion of the study findings, along with recommendations for future research. 79 Research Questions and Hypotheses As previously discussed, this study was guided by five research questions and hypotheses. These questions first sought to find if a student’s score on the Internet Addiction Test (IAT) was related to a student’s grade point average, and if the IAT score was predictive of success in college as measured by grade point average. Research questions two through five sought to better understand if there was a mean difference of IAT scores in the population between the variables of sex, research site, race, and student classification. The following section outlines both the research questions, accompanying directional and null hypotheses, and the specific statistical analysis utilized for each question. For additional discussion on the hypothesis anticipations, refer to Chapter Three. RQ1: “Is there a relationship between a student’s Internet Addiction Test (IAT) score and grade point average?” RQ1: H0/Null hypothesis: There is not a significant relationship between IAT score and grade point average. RQ1: Ha/Alternate hypothesis: Students’ scores on the IAT are significantly related to grade point average. RQ2: “Is there a mean difference among sex and IAT score?” RQ2: Ho/Null hypothesis: There is no significant difference of means of IAT score and sex. Ha/Alternate hypothesis: There is a significant difference of mean of IAT score and sex. RQ3: “Is there a mean difference among research site and IAT score?” 80 RQ3: Ho/Null hypothesis: There is no significant difference of means of IAT score and research site. Ha/Alternate hypothesis: There is a significant difference of mean of IAT score and research site. RQ4: “Is there a mean difference among race and IAT score?” RQ4: Ho/Null hypothesis: There is no significant difference of means of IAT score and race. Ha/Alternate hypothesis: There is a significant difference of mean IAT score and race. RQ5: “Is there a mean difference among classification and IAT score?” RQ5: Ho/Null hypothesis: There is no significant difference of means of IAT score and student classification. Ha/Alternate hypothesis: There is a significant difference of mean of IAT score and student classification. As discussed previously, this study used four phases of data analysis: A descriptive analysis provided an overview of the study respondents, Pearson’s r analysis was used for research question one to analyze the relationship of IAT score and grade point average, and a linear regression (OLS) analysis was utilized to show if the independent variable (IAT score) is predictive of the dependent variable (grade point average). Next, analysis of variance (ANOVA) was used for research questions two through five to determine if there was a difference of means among the demographic variables of sex, research site, race and student classification. The next section presents 83 identified as 4th year or more were 27.0% (n =187), 26.6% (n =184) identified as 1st year students, 24.8% (n =172) marked second year students, and 21.4% (n = 148) selected third year student. The study was conducted at three research sites. Table 4.2 depicts the descriptive data based on research site. Table 4.2: Descriptive Data of Study: Research Site Demographic Variables n Percentage Research Site: Institution Institution #1 (Regional Public) 412 59.5 Institution #2 (Regional Private) 162 23.4 Institution #3 (Regional Private) 116 16.8 Abstained 2 00.3 Total 692 Research Site: Institution Type Public Institution 412 59.5 Private Institution 278 40.2 Abstained 2 00.3 Total 692 Students participating at the regional, public institution, (institution #1) comprised 59.5% (n =412). Students participating from institution #2 made up 23.4% (n =162), and 16.8% of the participants (n =116) were from the other regional, private institution (institution #3). The breakdown of public to private students participating in the study was 59.5% (n =412) public institution to 40.2% (n =278) private institution. Two students (.3%) abstained from selecting an institution. 84 Statistical Analysis As previously outlined, the statistical analyses used to answer the research questions in this study were Pearson correlation (r), and linear regression (OLS) for research question one, and an analysis of variance (ANOVA) for research questions two through five. Assumptions for each analysis are presented with the accompanying research question and subsequent data analysis. This section presents each research question, accompanying assumptions, and the statistical results. Discussion and implications of the results of this study are found in Chapter Five. Research Question #1: “Is there a relationship between a student’s Internet Addiction Test (IAT) score and grade point average?” A Pearson r was used to address whether there was a relationship between IAT score and grade point average. The analysis found there was a significant, negative correlation between the two variables (r = -.320, n = 692, p = .000) for the two-tailed test (Table 4.3). Table 4.3: Pearson correlation GPA IAT Score GPA Pearson correlation 1.00 -.320** Sig. (2-tailed) .000 N 692 692 IAT Score Pearson correlation -.320** 1.00 Sig. (2-tailed) .000 N 692 692 ** Correlation is significant at the .01 level (2-tailed) Considering the results for research question number one, finding a significant, negative correlation between IAT score and grade point average (r = -.320, n = 692, p = .000) (Table 4.3), the null hypothesis, “There is not a significant correlation between IAT score and grade point average” was rejected, and the 85 directional hypothesis was retained, “Student’s scores on the IAT are significantly correlated with grade point average.” After finding a relationship was present between IAT score and grade point average, a linear regression analysis was chosen to address if a student’s IAT score predicted grade point average. Before conducting the regression, assumptions were analyzed. As outlined in Chapter Three, simple linear regression (OLS) has four assumptions (Lomax & Hahs-Vaughn, 2012; Nolan & Heinzen, 2012). The first assumption is that there is a linear relationship between the dependent and independent variables. A scatter plot is used to test the linear nature of the variables and is presented in figure 4.1. Figure 4.1: Scatter Plot of GPA and IAT score The scatter plot shows a negative, linear association between a student’s grade point average and IAT score. After finding a linear association between grade point average and IAT score, the second assumption of normality of residuals was tested. For the test of normality of the