Psychology Research Methods: Variables, Hypotheses, and Experimental Designs, Exercises of Psychology

Download Psychology Research Methods: Variables, Hypotheses, and Experimental Designs and more Exercises Psychology in PDF only on Docsity! Psychology Research Methods Summer Work WATCH WHAT | CAN MAKE PAVLOV Do. 2 SOON AS | DROOL, HE'VL SMILE AND WRITE IN HIS LITTLE Book.  2 What you need to know: The following comes directly from the AQA Psychology Specification. You can find all of this information on the AQA website Research Methods Students should demonstrate knowledge and understanding of the following research methods, scientific processes and techniques of data handling and analysis, be familiar with their use and be aware of their strengths and limitations: •• Experimental method. Types of experiment, laboratory and field experiments; natural and quasiexperiments. •• Observational techniques. Types of observation: naturalistic and controlled observation; covert and overt observation; participant and non-participant observation. •• Self-report techniques. Questionnaires; interviews, structured and unstructured. •• Correlations. Analysis of the relationship between co-variables. The difference between correlations and experiments. Scientific processes •• Aims: stating aims, the difference between aims and hypotheses. •• Hypotheses: directional and non-directional. •• Sampling: the difference between population and sample; sampling techniques including: random, systematic, stratified, opportunity and volunteer; implications of sampling techniques, including bias and generalisation. •• Pilot studies and the aims of piloting. •• Experimental designs: repeated measures, independent groups, matched pairs. •• Observational design: behavioural categories; event sampling; time sampling. •• Questionnaire construction, including use of open and closed questions; design of interviews. •• Variables: manipulation and control of variables, including independent, dependent, extraneous, confounding; operationalisation of variables. •• Control: random allocation and counterbalancing, randomisation and standardisation. •• Demand characteristics and investigator effects. •• Ethics, including the role of the British Psychological Society’s code of ethics; ethical issues in the design and conduct of psychological studies; dealing with ethical issues in research. •• The role of peer review in the scientific process. •• The implications of psychological research for the economy. Data handling and analysis •• Quantitative and qualitative data; the distinction between qualitative and quantitative data collection techniques. •• Primary and secondary data, including meta-analysis. •• Descriptive statistics: measures of central tendency – mean, median, mode; calculation of mean, median and mode; measures of dispersion; range and standard deviation; calculation of range; calculation of percentages; positive, negative and zero correlations. •• Presentation and display of quantitative data: graphs, tables, scattergrams, bar charts. •• Distributions: normal and skewed distributions; characteristics of normal and skewed distributions. •• Introduction to statistical testing; the sign test. 5 Activity 1 – Hypotheses Create a suitable hypothesis for the following aims: 1. An investigation into the effects of caffeine on reaction time. 2. An investigation into whether listening to Mozart affects memory 3. An investigation into whether gender affects performance in school. Variables When a hypothesis is created a Psychologist is usually attempting to identify a causal relationship between two factors, which we call variables. The way in which this is done is to change the variable we think is the cause, and measure whether this has an effect on the other variable. In psychology we talk about these variables as the independent (IV) and dependent variables (DV). IV = DV = For example, if we think that the grade you get in Psychology AS is affected by the amount of time you spend doing extended study, what would the IV and the DV be? IV: DV: In the examination you will need to be able to identify / create IVs and the DVs for a given scenario. Whenever you do this, you must check that you have gotten these variables the correct way around. The way to do this is to put your answers into the following sentence: (IV) affects (DV) 6 So, if we use the example from earlier, that the grade you get in Psychology AS is affected by the amount of time you spend doing extended study, which of the following sentences makes sense: 1) The grade you get in Psychology AS affects how much time you spend doing extended study OR 2) How much time you spend doing extended study affects the grade you get in Psychology AS Variables must be Operationalised – i.e. taken from being an abstract concept and made into something you can change (IV) or measure (DV) For example, if you take the aim “To investigate the effect of caffeine on Short term Memory” IV – Caffeine DV – Short Term Memory BUT these are not precise or testable, therefore they have not been operationalised. How can we vary Caffeine to make it something you can change? How could we make short term memory something you could measure? Activity 2 – Variables Identify the IV and DV from the following examples. Make sure you check you have them the right way round by using the sentence “IV affects DV”, and make sure they are Operationalised 1. Listening to music whilst completing extended study leads to worse grades than completing extended study in silence. IV: DV: 2. If you drink more than 3 caffeine based drinks a day you are more likely to be stressed. IV: DV: 3. People who are more attractive have an increased chance of being successful in job interviews IV: 7 DV: 4. You will become more intelligent if you study Psychology A level. IV: DV: 5. More intelligent people choose to study Psychology at A level than any other subject. IV: DV: Directional vs Non-directional Hypotheses Hypotheses can be directional (also known as one-tailed) or non- directional (also known as two-tailed). This refers to the prediction of the relationship between the IV and DV.  If the hypothesis tells you what effect the IV has on the DV (will it make it higher, lower, faster, slower etc) than it is called directional, because you can predict the direction of the effect of the IV on the DV.  If the hypothesis just suggests that there is going to be a difference due to the IV, but doesn’t tell you how the IV will affect it, it is a non-directional hypothesis. Directional – One tailed  Non-directional –Two tailed  or  Predicts the way the results will turn out Just says there will be a difference between groups Predicts which group will do better Do not know which will do better Could draw a rough graph from this prediction Could not draw a graph from the hypothesis Uses phrases such as;  Faster than...  Higher than...  Greater than...  More than... Uses phrases such as;  A difference between... 10 Activity 5 – Sampling Complete the table below, adding in an example of each sampling technique and a strength and weakness of each: Technique Description Advantages Disadvantages Random Volunteer / Opportunity Stratified Systematic 11 Experimental Design Once you have obtained your ppts, you then need to decide how you wish to use them. This is referred to as Experimental Design, and there are 3 main designs; independent groups, repeated measures and matched pairs / mixed design. Independent Groups You take your sample and split it into two separate groups using random allocation, e.g. flip a coin to decide who goes into which group. Group 1 complete condition 1 whilst group 2 complete condition 2. You then compare the results of the two groups to see the difference in their performance Advantages Weaknesses  Easy and quick to obtain and assign to separate conditions  Quick for ppts  Participant variables may affect the results – see below Participant Variables – these are the things which make us unique from each other. The problem with these is that, if we use random allocation then we could end up with an unequal balance of ppts in each group, e.g. all those with perfect eyesight in one condition, which may then make it seem that condition 1 performed better than condition 2 or vice versa, when really it has nothing to do with which condition they are in. Repeated Measures You take your sample and make them all complete condition 1 and then all complete condition 2. You then compare performance on condition 1 and 2 to see if there is a difference. Advantages Weaknesses  Eliminates participant variables, as the two groups are the same  Order effects may affect the results – See below  More likely to lead to demand characteristics, as ppt go through both conditions. – see single- blind as a way to overcome this Order effects – this is when performance in the second condition gets better due to practice (called the practice effect) or worse due to tiredness, or lack of motivation (called the fatigue effect). This could make it appear that the performance is better in one condition than the other, when really it is nothing to do with the condition itself, merely that they have been through the process twice. Order effects can be overcome by using counterbalancing. This is where although all ppts still complete all conditions, you split your sample in half, and then make group 1 complete condition 1 then condition 2, whilst group 2 do the opposite, i.e. do condition 2 then 1. This then cancels out any order effects, as if group 1 get worse in condition 2 due to tiredness, so do group 2 in condition 1. Note – there are some things that can never be repeated measures as they can’t be easily manipulated by the experimenter – e.g. gender, or to a lesser extent age (although this could be done through a longitudinal research method), nationality and so on. 12 Matched Pairs / Mixed design This is where you match each ppt in group 1 with a similar person in group 2 based on criteria which are relevant to the research – e.g. age in a test of memory. These must be relevant criteria though. There would be no point matching people on gender for a memory test, unless there was a suggestion that gender affects memory. Then group 1 completes condition 1, group 2 condition 2, and the results are compared. Advantages Weaknesses  Minimises participant variables, as the groups of people are similar  No order effects, as ppts only go through one condition.  May be difficult to match people on all relevant variables.  May be some variables which are difficult to measure to therefore match people on.  There will still be some participant variability. Activity 6 – Experimental Design Use the information above to complete explain how you would use each experimental design to test the following hypothesis: “Those who listen to Mozart whilst learning a list of 40 words will remember a greater number of words when tested than those who learn the words in silence.” Independent Groups Repeated Measures Matched Pairs / Mixed Design 15 Complete definitions of each in the space below: Validity: Reliability: Why is it important to have validity in research? Why is it important to have reliability in research? Validity can be further split into different types. The main 2 that we are interested in are; Internal Validity External Validity This is whether we have been able to establish cause and effect in the research, or whether the results may be due to confounding or extraneous variables This is whether our conclusions can be accurately applied to other people and other settings beyond the research setting. A related issue here is mundane realism – how like real, everyday life was the task the ppts completed? 16 Types of Experiment Whilst all experiments involve the measurement of the DV after change of the IV, there are several different ways in which may be conducted Laboratory Experiments A Laboratory should not be considered as a specific room, but instead a controlled environment. Anywhere could be your laboratory, provided you are able to have a high level of control over the variables in that place. In a laboratory the IV would be directly manipulated by the researcher. This high level of control is excellent in establishing cause and effect, as any extraneous variables will be easily controlled, and precise manipulation and measurement of the IV and DV are enabled, however, it is unlikely that a laboratory would yield research which is high in realism, as it is so artificial and likely to cause demand characteristics. Field experiments The field simply refers to a natural habitat, so anywhere the ppt would have been at that time without the research taking place. The researcher still manipulates the IV, but the environment is not controlled, which could lead to extraneous variables having an effect on the DV. A field experiment is likely to have greater realism and lower demand characteristics than a laboratory experiment. Natural Experiments The word natural refers only to the way in which the IV is varied. If it is something which the experimenter cannot directly control, then this termed a natural experiment. A natural experiment could take place in a laboratory or in the field, as deeming something a natural experiment has nothing to do with the setting in which the research will be carried out. For example, if I were to compare the memory of students from grammar school and high school, I could not randomly decide who should go to which school purely for the purposes of my research, but I could still use a laboratory to investigate the effects. A further type of experiment is known as a quasi experiment. Some will group quasi and natural experiments together, although there is a subtle difference between these. In a quasi experiment the IV will be naturally occurring, such as age or gender, whilst in a natural experiment, the IV could be manipulated by someone else, but not the researcher, e.g. the Island of St Helena had TV introduced by television networks, but researchers were still able to investigate the effect using a repeated measures design. Complete the summary table on the next page to identify the advantages and weaknesses of these types of experiments: 17 You need to know at least two advantages and two weaknesses of each method. Method Description/use Advantages Weaknesses Laboratory Experiment Field Experiment Natural Experiment 20 Strictly speaking a correlational study is not an experiment – cause and effect are not being predicted. This is also true as in correlations, often neither variable is manipulated by the experimenter. Many correlation techniques calculate a correlation coefficient, basically a measurement of the strength of the correlation, a statistic which has a value on a scale between +1 and –1. negative correlation (-1 = perfect negative) Zero correlation positive correlation (+1 = perfect positive) means that high scores on one variable are associated with low scores on another variable There is no relationship between the two variables means that high scores on one variable are associated with high scores on another variable The results can be plotted on scattergrams, which always have a numerical scale for each axis, and should be labelled clearly, with a title for the graph. Activity 9 – Correlations Describe the correlations between the variables plotted on these graphs: Advantages Weaknesses 21  Gives a numeric value to the strength of a relationships between variables  It can be used in natural settings as well as in the laboratory; such research is known as field research.  The correlational method is often highly efficient and can yield a large amount of data relatively quickly.  It can be extended to include many different variables at once.  It can be used where experimental manipulation would be unethical or impossible.  It can indicate a trend which can then be used to guide experimental research  it always raises questions of chicken and egg situations – does smoking cigarettes makes one’s attitude to advertising to be more lenient, or does one’s lenient attitude to advertising cause one to smoke more cigarettes?  the relationship may be due to other extraneous variables, for example height and IQ might be linked because diet influences both.  The findings it yields are generally not conclusive with respect to cause-and- effect relationships. That is, the fact that two variables are correlated, even highly correlated, does not guarantee that there is a causal link between them-that changes in one cause changes in the other. Rather, in many cases, the fact that two variables are correlated simply reflects the fact that changes in both are caused by a third variable.  measurement of non-linear relationships not possible (e.g. attention span and time of day – relationship may be positive early in day and negative late in day – a non-linear relationship) 22 Quantitative Data Analysis and presentation Data is the results from research. Rather than presenting all of this in its raw form in your report we use data analysis, descriptive and inferential statistics to summarise these results. This means that anybody reading the report of the study will have a concise summary of the results and conclusions can be reached. Descriptive statistics Descriptive stats allow research data to be described and presented. It is not helpful to the reader to be given raw data of a study but it is important that they have a summary of that data. This may take the form of: o A table o A graph o Numerical average Measures of central tendency 1. Mean – when all scores in a group are added together and the total is divided by the number of scores. Eg. the results from a test (marked out of 50) 36 39 21 18 32 30 = 176 176 / 6 = 36 Mean = 36 2. Median – this is the central value in a set of scores after they have been put in rank order: Eg. 95 109 121 130 140 Median = 121 If there is an even number of scores take the mean of the two central values: Eg. 95 109 121 135 140 180 121 + 135 = 256 / 2 = 128 3. Mode - this is the most commonly occurring value in a set of scores: Eg. 1 2 3 3 3 4 4 4 5 5 5 5 5 5 5 5 6 6 6 7 Mode = 5 This handy rhyme should help you remember which of the measures of central tendency is which: Hey diddle diddle, the Median’s the middle. You add and divide for the Mean The Mode is the number most commonly found, The Range is the difference between. 25 Presentation of quantitative data Tables The first thing a psychologist will do with the results they have collected is create a table of their results. In your exams you could be presented with a table and have to interpret it. The first thing you need to know about a table is what experimental design was used. This should be outlined in the blurb given before the graph. You must know this so that you can state findings from the table easily. Independent Groups Condition A Condition B 4 7 6 6 3 9 9 6 2 2 5 8 3 9 Repeated Measures Ppt Condition A Condition B 1 4 7 2 6 6 3 3 9 4 9 6 5 2 2 6 5 8 7 3 9 As you can see, in the IG table, the rows that people are in are irrelevant, whereas in the RMs table, the rows people are in are very important, as this is one ppts performance in two different conditions. 26 Graphs Graphs are used to display data in a form which is easy to read. You will be familiar with different types of graphs but it is important to know when to use each type and what they mean. Bar charts – these use bars which do not touch and can be used with a variety of types of data, and will usually be used to show a specified measure of central tendency, such as the mean, for the conditions in a piece of research. Scattergrams / Scatterplots / Scattergraph – This is where two sets of continuous numerical data are plotted against each other, where a line of best fit can then be used to represent the data and identify whether there is a correlation between the two variables on the axes. Graphs must have:  Clearly labelled axes which include units / maximum values  Both axes starting at zero – otherwise the data will be mis-represented  A title which describes the graph  The IV on the X axis and the DV on the Y axis – this rule does not apply for correlations Activity 12 - Graphs Produce an appropriate graph to display the following data: 1. Results of a study into how age affects types of play: Play Age (years) 1 2 3 Solitary 16 8 4 Parallel 3 9 7 Co-operative 1 3 9 2. Scores from Ps who were asked to rate speakers on intelligence on a scale 0-5 (where 5 was very intelligent) when speakers had Northern and Southern English accents. Rating Southern Northern 1 0 0 2 5 6 3 12 23 4 26 21 5 21 10 27 3. Results from a correlation study to see if practice on a driving test improves performance: Number of attempts Points awarded 1 27 2 54 3 78 4 105 5 120 6 149 4. Ps who sleep more than 7 hours per night (on average) in one year will gain higher marks in the final A’Level exam than Ps who sleep less than 7 hours per night (on average) in one year. The results for this study are as follows: P Hours of sleep (average) Exam mark 1 8.5 73 2 5.8 52 3 5 35 4 6.1 61 5 7.4 66 6 6.9 70 7 7.4 65 8 6.6 56 9 7.5 71 10 8.9 79