Frequency Distributions and Central Tendency: A Statistical Analysis Primer, Exams of Psychology

Download Frequency Distributions and Central Tendency: A Statistical Analysis Primer and more Exams Psychology in PDF only on Docsity! intro to statistics in psychology midterm CHAPTER ONE - ... cases - are the subjects of a study or objects on whom variables are being measured variables - are characteristics measured by researchers statistic - if a number characterizes a sample correlation design pg 7 - variables are simply measured or observed, as they naturally occur. components of correlation design - predictor: variable that comes first chronologically or is believed to drive a relationship criterion: outcome variable, it is where the predictor variable influence is observed confounding variable - 3rd variable tht has an impact on both variables your interested in pros of correlation design - allows to study relationships between variables (natural environment) cons of correlation design - doesn't allow conclusion about cause and effect experimental design pg 9 - variables are manipulated or controlled by the researcher. components: independent variable(IV) controlled by the experimenter dependent variable(DV) effect of the manipulation components of experimental design - independent variable (IV): controlled by the experimenter dependent variable (DV): effect of the manipulation remember by ICED: independent- controlled effect- dependent pros of experimental design - have control and can be more confident cons of experimental design - not always realistic quasi-experimental design pg 11 - study where cases are classified to different groups by some characteristic they already possess. components of quasi-experimental design - GROUPING VARIABLE: variable that defines the group in which a participant belongs (natural choice of participant) DEPENDENT VARIABLE: variable where the influence of the grouping variable is measured parameter vs. statistic - parameter looks at a population statistic looks at the sample descriptive vs. inferential statistic - order of operations: - PEMDAS parenthesis exponents multiplication division addition subtraction rounding rules: - --round to 2 decimal places for final answer --don't round until the end --carry 4 decimal places as you work --if the unrounded number is exactly between 2 rounding options round up CHAPTER 2 - ... frequency distribution - count of how often different values of a variable occur in a set of data --2 types: grouped and ungrouped grouped frequency distribution - --condenses info by creating intervals --helpful when large number of possible values ungrouped frequency distribution - displays all possible values -- provides detail cumulative frequency - (fc) tells us how many cases in a dataset have a given value or a lower value --top row should =N ?? -- calculated by adding up all the frequencies at or below the given row (table on pg 42) how are frequency distributions organized? - upside down, with the biggest value the variable can take on the top row (should be the same as the total number of cases (N) percentage - way of transforming scores to put them in context --the % column takes the info from the cumulative frequency column and turns it into % percentage formula - %= f/N x 100 provided on exam cumulative % - the % of cases with scores at or below a given level -- should equal 100% on the top row cumulative percentage formula - %c= Fc/N x 100 provided on exam discrete numbers - answers the question "how many?" --take whole number values and have no in between or fractional values --norminal and ordinal level continuous numbers - answers the question "how much?" --can take on values between whole numbers --always interval or ratio level difference between discrete and continuous - discrete can only be a whole number, for example amount of people you can't have 2.5 continuous can take on any value for example height used to demonstrate the frequency with which the different values of discrete values occur --used for discrete data --height of the bars indicates the frequency - bar graph a graphic display of a frequency distribution for continuous data --used for continuous data --the bars touch each other bc it has numbers that take on decimal places - histogram displays the frequency distribution of a continuous variable as a line --sometimes called a line graph interquartile range (IQR) - distance covered by themiddle 50% of scores --a single number variance - mean of squared deviation scores standard deviation - square root of the variance formula for population variance - provided on exam ó^2= E(X-U)^2/ N E-sum X-raw score U-population mean N-number of cases in the population formula for population stdev - provided on exam ó=squareroot of population variance formula for sample variance - provided on exam s^2= E(X-M)^2/ N-1 formula for sample stdev - s= squareroot of s^2 (sample variance) CHAPTER 4 - ... standard scores (z scores) - raw score expressed in terms of how many standard deviations it falls away from the mean why and how are z scores used? - --compare different metrics --how rare a case is FORMULA: z= X-M/ s (provided on exam) normal distribution - bell shaped curve that is defined by the percentage of cases that fall in specific areas under the curve normal distribution (curve) characteristics - --symmetrical --midpoint is the mean, median, and mode --as the scores move away from the midpoint the frequency of their occurrence decreases symmetrically probability - the number of ways a specific outcome can occur, divided by the possible number of outcomes formula for probability - not provided on exam p(A)= # of ways A can occur/ total # of ways CHAPTER 5 - ... representative sample - all the attributes in the population are in the sample in the same proportions by which they are present in the population why should a sample be representative? - the results of the study can be generalized fro m the sample back to the population convenience sample - cases within the population of interest that are easily gathered -- unlikely to be truly representative random sample - all cases in the population have an equal chance of being selected self selection bias - subjects who agree to participate in a study differ from those who choose not to participate why is self selection bias a problem - if the ppl who chose to participate differ in some way from those who chose not to participate then the sample is no longer representative of the population example: if the topic is sex some people will not respond because this is a personal topic, some who chose to respond could have a problem with sex or could be addicted sampling error - discrepancies between the sample values and population values due to random factors why is sampling error a problem -