Methodology and Program Evaluation in Psychology
Methodology and Program Evaluation in Psychology PSYC 611
Popular in Course
Popular in Psychlogy
This 23 page Class Notes was uploaded by Ozella Cassin on Monday October 19, 2015. The Class Notes belongs to PSYC 611 at Radford University taught by Jeffery Aspelmeier in Fall. Since its upload, it has received 29 views. For similar materials see /class/224720/psyc-611-radford-university in Psychlogy at Radford University.
Reviews for Methodology and Program Evaluation in Psychology
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 10/19/15
Stats Review Data Analysis as a Decision Making Process I Levels of Measurement NOIR See Whitley 2001 pp 350351 for details Nominal Categories with Names Yes vs No don t ask sometimes vs never Sex Religion Relationship Status Political Af liation Experimental Group Membership Control Group Manipulation Group Comparison Group Numbers represent groups 1 Female 2 Male Order is arbitrary Ordinal Nominal Categories with a Logical Order Class Rank Height from tallest to shortest Responses on a Numerical Rating Scale 1 strongly agree 7 Strongly disagree Intervals between numbers is not standard from unit to unit Interval Numerical scales with logically ordered units that are equidistant but the zero is arti cial Eg Temperature in Centigrade and Fahrenheit Zero does not represent an absence of temperature Time of day no 0 o clock Calendar Dates Calendar Years Ratio Numerical scales with logically ordered units that are equidistant and have a true Zero the zero represents a lack of that which is being measured E g Elapsed Time Temperature in Kelvin 0 degrees Kelvin 27315 degrees celcius Length Mass Because it uses a true zero numerical values can be used to de ne ratios 5 inches is ve times more length than 1 inch 10 inches is twice as long as 5 inches Ratio Level Measures are Quite Rare in Psychology Continuous Vs Discrete Variables Discrete Variables Mutually ExclusiveExhaustive Numerical Categories that can t be broken down in to ner units e g if sex is represented by lmale and 2female there is no 15 All Nominal and Ordinal Variables are Discrete However many Ordinal variables will be treated as continuous eg the Numerical rating scales are often averaged to form a single score which is treated as continuous Continuous Variables Numerical systems where there are an infinite number of possible points between each unit Also the measurements can be broken down into ner units eg elapsed time Years Months Days Hours Minutes Seconds Milliseconds Nanoseconds etc II Choosing your Statistics Knowing which statistic to use to test the relationship between each variable depends on the type of data you have and sometimes the type of question you want to answer Note This is often discussed with respect to the issue of statistical validity and there are different camps regarding the Appropriateness of certain levels of measurement for speci c statistics Michell 1986 Also given specific conditions most parametric statistics can hand nonparametric data A Single Discrete Variable Goodness of Fit X2 Allows us to test whether the group frequencies differ from chance patterns base rate frequencies the frequency instances naturally occur in the environment df k l where k number of groups 2 Of Ei2 Z Z T where Oi observed frequency for each separate group Ei expected group frequencies based on chance Statistical Hypotheses Ho Of Ef Ha Of Ef Example Research 3 Question Does number of people who say they like cheezy poofs Yes 1 vs those who do not like cheezy poofs No 0 differ signi cantly from the number expected by chance alone B Discrete X Discrete Pearson s X2 AKA Test of Independence Allows us to test whether the cross tabulation pattern of two nominal variables differs from the patterns expected by chance If one variable is ordinal then I or F are norrnall used df RlCl where R ofrows amp C ofcolurnns Z 2 7 0 1 7 Ev 7 Z 7 2 E where 1 7 the different groups for Vauable l j the different groups for Variable 2 RC E where Ri Row total of rowi J N Cj Column total of column j Statistical Hypotheses Ho Of Ef Ha Of Ef Example Research 3 Question Does the number of people who think they are Eric Cartman yes 1 no 0 relative to whether or not they eat cheezy poofs signi cantly differ from the frequencies that are expected by chance alone Note a signi cant Pearson s chi square will not tell you which cells are different The crosstabulation matrix must be examine to determine this Eyeballing the standardized corrected residuals seems to be the most useful Limitations on X2 1 Responses must be independent and mutually exclusive and exhaustive Each case from the sample should t into one and only one cell of the cross tab matrix 2 Low expected Frequencies limit the validity of X2 If df l eg 2x2 matrix then no expected equency can be less than 5 Also lfdf 2 all expected frequencies should exceed 2 If d3 or greater then all expected frequencies except one should be 5 or greater and the one cell needs to have an expected equency of l or greater Phi Coefficient if 2X2 matrix correlation coef cient that estimates the strength of the relationship between two dichotomous nominal variables Note Phi can not estimate the direction eg positive linear vs negative linear of the relationship between 2 nominal variables because the numerical values are arbitrary direction is meaningless 39s correlation coef cient can be calculated exactlylike Pearson s r below or can be estimated using the XZ statistic Thus any XZ canbe convertedto Phi or Phi canbe converted to X2 signi cance should be determined using XZ tables 2 7 and ZZ ZN Statistical Hypotheses Ho Phi 0 Ha Phi 0 Example Research 3 Question What is the strength of the relationship between whether one thinks they are Eric Cartrnan or not Yes 1 No 0 and whether one eats cheezy poofs Yes 1 No 0 Cramer s Phi or V If 2X3 matrix or larger correlation coef cient that estimates the strength of the relationship between two discrete nominal variables correcting for the in uence of the number of groups Signi cance should be determined using XZ tables Note Phi can not estimate the direction positive vs negative of the relationship between 2 nominal variables because the numerical values are arbitrary direction is meaning ess mmw where k is the smaller ofR or C or when r c k r c N k 7 1 Statistical Hypotheses Ho Phi 0 Ha Plu 0 Example Research 3 Question What is the strength of the relationship between whether one thinks they are Eric Cartrnan or not Yes 1 No 0 and whether one prefers cheezy poofs 1 HooHoo Dillies 2 or coacoa yumyum s 3 Ifboth discrete variables are Ordinal Ranked Data Spearman s Rho r Correlation coef cient that estimates the strength and direction of the relationship between two ordinal variables Calculated using Pearson s r formula However the r table can t be used to estimate signi cance For N gt 10 and N lt 28 convert to a Z score IfZ greater than 196 then alpha lt 05 IfZ greater than 285 then alpha lt 01 For smaller and larger N respectively special tables will need to be obtained rs 22N 5 9N N 7 1 Statistical Hypotheses Ho Rho 0 a E Haero0 xam le Research uestion What is the strength and direction of the relationship between Eric Cartman s rankings of20 participant s suitability as a mate range 120 and participants rankings with respect to how many Cheezy Poofs they eat perweek range 1 20 Kendall s Tau 1 This statistic provides an alternative to Rho that is easier to calculate signi cance for by hand Rather than being a variation of Pearson s r Tau is based in the number of inversions between the ranked items For example the following data set contains 4 inversions out of the 10 comparisons Subi ect 1 2 Q i i E 1 2 Var 1 l I i 4 a 6 7 10 Var 2 6 7 39 10 Inversion Inversion Inversion Inversron 21 NN7 1 139 1 where I number of inversions Often Tau is used as a measure of interrater agreement though Rho can also be used for such oses Signi cance regardless of the number of subjects is tested using the same Z test as for 0 only substitute E for r Use the same Critical Values of Z If Z greater than 196 then alpha lt 051fZ greater than 285 then alpha lt 01 Statistical Hypotheses Ho Tau 0 Ha Tau 0 Example Research 3 Question What is the strength and direction of the H relationship between Eric Cartman s rankings of 20 participant s suitability as a mate range 120 and participants rankings with respect to how many 4 Cheezy Poofs they eat perweek range 1 20 C Discrete X Continuous 1f Discrete Variable is Dichotomous only 2 levels ztest compares a single sample mean to the population mean when the population standard deviation is known 2 7 y Z where u amp o population mean and standard dev1ation respect1ve1y W Signi cance Use the Critical Values on IfZ greater than 196 then alpha lt 05 1f Z greater than 285 then alpha lt 01 Statistical Hypotheses Ho Group 1 mean Population Mean Ha Group 1 mean Population Mean Example Research 5 Question Does the average number of Cheezy Poofs eaten per day by students enroled in Graduate Research Statistics range 0600 differ from the average number of Cheezy Poofs eaten per day in the general population of Graduate Students t test Signi cance If I obtained exceeds the t critical see any ttab1e for critical values for a given df at the 05 alpha level then the groups are signi cantly different Single Sample t test compare a sample mean to a population mean when the only the sample standard deviation is known f n t where s sample standard deviation J Statistical Hypotheses Ho Group 1 mean Population Mean Ha Group 1 mean Population Mean Example Research 5 Question Does the average number of Cheezy Poofs eaten per day by students enroled in Graduate Research Statistics range 0600 differ from the average number of Cheezy Poofs eaten per day in the United States Independent Sample t test compare two the means of two unrelated groups df n2 t Y1 Y2 n1 1s12 n2 Us E 1 1 j 7 7 n1 n2 2 n1 n2 Statistical Hypotheses Ho x1 gt12 Ha gt lt1 gt lt2 Example Research Question Does a randomly assigned group exposed to 37 hours of South Park group l reruns report significantly more positive or negative attitudes toward Research Methods measured using a 5 item questionnaire employing a 7 point rating scale item averages range from 17 compared to a randomly assigned comparison condition exposed to 37 hours of Sally Struthers Feed the Children commercials group 2 Repeated Measure Matched Sample t test Repeated Measure test the signi cance of the averaged difference in scores between time 1 and time 2 Matched Sample compare averaged difference in scores between group 1 and group 2 when the subjects from each group have been matched on some variable e g age intelligence etc df n 1 X11 X12 t n dif 2 Z X X 2th X12 lt lt 2 1 n V n 1 72 Statistical Hypotheses Ho in in Ha in gtltt2 Example Research Question After being exposed to 37 hours of South Park reruns Time 2 do participants report significantly more positive or negative attitudes toward Research Methods measured using a 5 item questionnaire employing a 7 point rating scale item averages range from 17 compared to their preexposure scores Time l Biserial vs Point Biserial Arti cial Dichotomies vs Natural Dichotomies Point Biserial Correlation Coef cient rpb Same formula as Person s r see below only one variable is a natural dichotomy often designated by 0 amp 1 This correlation coefficient will indicate the strength of the relationship between category membership and the continuous score Note that like Phi the direction of the relationship positive vs negative is arbitrarily based on the numerical labels assigned to the groups Examination of the means is necessary to determine the direction of the group differences df n 2 Statistical Hypotheses Ho rpb 0 Ha r 0 pb Example Research Question What is the strength of the relationship between being randomly assigned to a group exposed to 37 hours of South Park reruns group l vs being randomly assigned to a comparison condition exposed to 37 hours of Sally Struthers Feed the Children commercials group 0 and selfreport attitudes toward Research Methods measured using a 5 item questionnaire employing a 7 point rating scale item averages range from 17 Biserial Correlation Coef cient rb Used when a Dichotomy is developed from a continuous variable e g mean split or median split methods Groups are often designated by 0 amp l rb is estimated from a normal Pearson s r see below This statistic will tell you the strength of the association between category membership based on an arti cial dichotomy and a continuous score Again direction of the relationship will be arbitrary depending on the numerical category labels however since they are based on continuous scores the numerical label with greater value should be given to the upper end of the continuum making interpretation easier df n 2 0 o rpearson A belowcp A abovecp r b Xe 0 where ch the raw score cut point raw score used to split distribution into 2 groups Statistical Hypotheses Ho rb 0 Ha rb 0 Example Research Question What is the strength and direction of the relationship between watching more than 10 hrs per week of South Park Group 1 vs watching 10 or fewer hours of South Park per week group 0 and selfreport attitudes toward Research Methods measured using a 5 item questionnaire employing a 7 point rating scale item averages range from 17 where 10 hrs per week is the mean of self reported South Park viewing habits If Discrete Variable has more than 2 Levels more than 2 groups One Way ANOVA Anova tests whether 2 or more group means are signi cantly different When using 2 groups F t2 See ANOVA handout for details on One Way Anova Statistical Hypotheses Ho Mean grpl Mean grp 2 Mean grpj forj groups Ha At least one group mean signi cantly different from one other group mean Example Research Question Are there any signi cant difference between three randomly assigned groups l exposed to 37 hours of South Park Episodes 2 exposed to 37 hours of Sally Struthers Feed the Children commercials amp 3 no TV control condition with respect to their selfreport attitudes toward Research Methods measured using a 5 item questionnaire employing a 7 point rating scale item averages range from 17 If Discrete X Discrete X Continuous Where both discrete variables are predictors Two Way ANOVA If we have 2 independent variables or 1 IV and a Blocking Variable then Two Way Anova Or Factoral ANOVA is called for Again this will tell us if one group mean or matrix cell mean is signi cantly different from one other group mean or cell mean Also Factoral Anova can handle More than 2 IV s and or Blocking Variables See Two Way Anova Handout for Details Moderation when we nd a significant interaction between two predictor variables we can say that one predictor moderates the relationship between the other predictor and the outcome DV Our decision about which predictor is the IV and which is the Moderating Variable is based on our theoretical perspective Statistical Hypotheses The two way ANOVA actually tests several hypotheses at once 1 Main Effects IVl Ho Mean Group 1 Mean Group i for i groups Ha At least one group mean significantly different from one other group mean IV2 Ho Mean Group 1 Mean Group j forj groups Ha At least one group mean significantly different from one other group mean 2 Interaction effects moderation effects 1 Ho Mean grpll Mean grp 21 Mean grp12 Mean grp ij fori and j groups Ha At least one group mean significantly different from one other group mean Example Research Question Do males sex 0 have significantly more positive or negative attitudes toward Research Methods measured using a 5 item questionnaire employing a 7 point rating scale item averages range from 17 compared to females sex 1 Also does a randomly assigned group exposed to 37 hours of South Park group l reruns report significantly more positive or negative attitudes toward Research Methods measured using a 5 item questionnaire employing a 7 point rating scale item averages range from 17 compared to a randomly assigned comparison condition exposed to 37 hours of Sally Struthers Feed the Children commercials group 2 Does participant sex m 0 f 1 moderate the relationship between south park exposure IV random assignment to condition 1 exposed to 37 hours of South Park Episodes 2 exposed to 37 hours of Sally Struthers Feed the Children commercials and attitude toward Research Methods measured using a 5 item questionnaire employing a 7 point rating scale item averages range from 17 ANCOVA Analysis of Covariance Sometimes we want to remove the effects of a third variable Covariate Can be continuous or discrete We may want to remove the effects of a nuisance variable that is correlated with our main variables of interest or we may want to test a mediational hypothesis that the 3rd variable explains the relationship between the IV and DV that is the 3rd variable accounts for all the shared variance between the IV amp DV see a good stats book for details Statistical Hypotheses Ho Mean Groupl Mean Groupj for j groups Ha At least one group mean significantly different from one other group mean Example Research Question Do males Sex 0 have significantly more positive or negative attitudes toward Cheezy Poofs measured using a 6 item questionnaire employing a 5 point rating scale item averages range from 15 compared to females Sex 1 after the effects associated with IQ range 70150 are removed NOTE Multiple Regression Anything Anova and Ancova can do Multiple regression can do as well through the use of dummy coding effects coding and contrast coding D Continuous X Continuous 1 Pearson s r Allows us to test the strength of the association between two continuous variables It represents a ratio of the Covariance variance shared by two variables and the total variance covariance unique variance df n 2 ZXYZ EXXEY lzX2 EX 2J2Y2 2W 1 2 X YYJZ Y l72 n n Statistical Hypotheses Ho r 0 Ha r 0 Example Research Question What is the strength and direction of the association between the number of Vienna Sausages that a person eats in 24 hr period measured in grams range 0900 and the amount of time they spend doing impersonations of south park characters in that same 24 hr period measured in minutes range 0 1440 The Coef cient of Determination r represents a ratio of covariance to total variance but if we want to know how much variance in the DV is explained by variance in the IV then we can square r to nd out NOTE for all of the correlation based statistics above Rho Tau Biserial PointBiserial if the sample size exceeds 30 then the Pearson r is robust enough that it can will provide a reasonable approximation of any of these statistics without significantly in ating the Type I error rate Signi cance of r signi cance tables for r are available in most stats books If r obtained exceeds the critical value of r for a given df at the 05 alpha level then there is a signi cant association between the variables of interest If r tables are not available you can use ttables as r tables are based on t conversions t l n 2 l r2 2 Partial Correlation pr Tests the association between two continuous variables with the effects of a third variable removed We may want to remove the effects of a nuisance variable that is correlated with our main variables of interest or we may want to test a mediational hypothesis that the 3rd variable explains the relationship between the IV and DV that is the 3rd variable accounts for all the shared variance between the IV amp DV see a good stats book for details Note The covariate does not necessarily have to be continuous especially if your N is larger than 30 Statistical Hypotheses Ho pr 0 Ha pr 0 Example Research Question What is the strength and direction of the association between the number of Vienna Sausages that a person eats in 24 hr period measured in grams range 0900 and the amount of time they spend doing impersonations of south park characters in that same 24 hr period measured in minutes range 0 1440 when the variance in attitudes toward Vienna Sausages measure using 4 item measure employing a 7 point numerical rating scale item averages rang from 1 to 7 is removed 3 Regression When you have a single variable regression is essentially the same as correlation Instead of r you calculate for b beta where b re ects the slope of line that best ts the data leastsquares regression coef cient Strength of association is represented as the change in Y DV that results from a 1 unit change in X IV Equation for a Straight Line LeastSquares Regression Y a bX Where Y Dependent Variable seen as a function of or predicted by the independent variable X X Independent Variable the dimension or characteristic that is seen as the determinant or cause of the dependent variable Y b Slope ofthe Line rise or drop divided by run a Y z39ntercept where the value of X 0 and the line intercepts the Y axis Formula for Calculating a and b 2XY EXXZY b azf bf EX2 2X2 n Once a and b are known they can be plugged into the regression line formula Y a bX and the predicted value of Y can be estimated for any value of X Example Research Question For a given slope b of 6300 and a Yintercept a of 32 then how much time would a person be expected to sped impersonating Southpark characters measured in minutes range 01440 in 24 hour period after eating 300 grams of Vienna Sausages E Multiple Continuous Independent Variables and Single Continuous Dependent Variable Multiple Regression Like Factorial Anova Multiple Regression can deal with multiple Independent Variables Also as a form of regression multiple regression can be used for prediction predicting Y based on the values of the IVs R The Multiple Correlation Coef cient indicates the total association between the Predictors IVs and the Criterion DV R2 Squared Multiple Correlation Coef cient indicates the of the variance in the DV that is accounted for by the IVs F used to test the significance ofR b The multiple regression coef cient indicates the increase in Y resulting from a 1 unit increase in Y when all other IVs are held at the constant of 0 That is it tells us about the unique effect a given predictor has on the criterion 6 Beta The standardized multiple regression coef cient same as b only the units are standardized in Zscore units t bstandard error for b used to test the signi cance of the regression coef cient The research questions that you can ask with Multiple regression are quite exible Single Step identi es the unique association of each predictor with a criterion Hierarchical Regression Mulitple Steps identi es the unique contribution of a single variable or group of variables to the multiple correlation coef cient RZA Mediation Analyses Because A third variable accounts forExplains the relationship between X and Y Why is X related to Y because of Z This is the ultimate goal of Science Moderation Analyses Interaction effects It Depends A third variable in uences the strength andor direction of the relationship between an IV and DV What in uence does X have on Y It depends on Z F Multiple Dependent Variables 1 Discrete IVs and Multiple Continuous DV s MANOVA Multivariate Analysis of Variance When you have multiple continuous outcome variables that re ect a related set of constructs and you want to test the association with l or more discrete IV s you can use Multiple Analysis of Variance Returns a single F that indicates whether the IV or IVs are signi cantly associated with the DVs as a group This is most useful for keeping the Type I error rate down when conducting multiple analyses If it is signi cant then it is usually followed up with Univariate tests assessing one dependent variable at a time see a good stats book for details MANCOVA Manova with a Covariate a variable that is having it s in uence removed from the test see a good stats book for details 2 One or more Continuous Independent Variables and Multiple Continuous Dependent Variables Canonical Correlation 0r Set Correlation Returns a single correlation coef cient that is the best tting correlation between set 1 IVs and set 2 DVs determined through multiple iterations Path Analysis Structural Equation Modeling TheoryModel Testing Procedures Allows you to determine the degree to which causal relationship predicted by theory t with the data Goodness of Fit which is expressed as a chisquare and other t indicies Path Analysis Causal Modeling only considers Manifest Variables which are directly measured variables Sample Palll Model Cllcczy Poet39s Liking C l39lmml 9 Appearance Vienna Sausage Consumption Structural Equau39on Modeling Latent Variable Models include Latent Variables which are not measured directly For example SES is not determined by a single indicator It is alatent variable made up ofmanifest variables like income education job prestige and obtained wealth Sample Structural Model Chcezy Puufs Eating Cheezy Poofs Liking Checzy Poofs 11 rlng Chcezy Poofs Evaluation G Continuous predictors and Categorical Dependent Variables Logistical Regression Allows you to ask avariety ofresearch questions with minimal statistical assumptions eg assumptions regardin normal distributions The most important ofwhich would be Strength ofAssociation between future cases Logistic regression can handle multiple predictors that either continuous or discrete and combinations ofboth Measurement Strategies Reliability A Classical Testing Theory 1 Reliability The Consistency of a Measure Does it always measure the same thing 2 Classical Testing Theory 2 Assumptions a All measures suffer from inaccuracies Unreliability random vs systematic b More Measurements More Reliability A Classical Testing Theory 3 Modeling Reliability 39 Xi ti ei Where Xi Subject Score 2 i True Score ei Error More Generally 2 2 2 O x 0 true score 0 error A Classical Testing Theory 3 Estimating Reliability Reliability Coefficient ratio of true score variance to total variance cf 0 a Reh39abiily rm 2 2 39 7X CYX If oX Gem Then Reliability If Gem 0 Then Reliability If oX otmescore Then Reliability 1 rXX ratio of error variance to total vanance B 9 Estimating Reliability Multiple Presentations of a Test TestRetest Reliability Give Same Test at two Time points X1 amp X2 Correlate Results High Positive Correlations Good Reliability Moderate Reliability 50 across 13 months Practice Effects State vs Trait Constructs B 1 Estimating Reliability Multiple Presentations of a Test b Parallel Forms Present 2 Conceptually Identical Forms Items differ Control Practice Effects Nonequivalence vs unreliability B Estimating Reliability 1 Multiple Presentations ofa Test c Reliability Coef cient rxx 02 r r 1 2 x 039 Numerator Covariance Between Time 1 and Time 2 or Form 1 and Form 2 Denominator pooled variance for T1 and T2 Form 1 and Form 2 B Estimating Reliability 1 Multiple Presentations ofa Test c Reliability Coef cient De nitional Formula 7 2X 7 X1X3 7 X3 2X X13 2X2 A72 Computational Formula ZXTXZ X2 2X1X2 2 X22 X1 7 r ZXf I7 B Estimating Reliability 2 Internal Consistency Item Covariation Single Test with Multiple Items a Split Half Reliability Split test into equivalent halves eg oddeven random split first V2 second V2 essentially Parallel Forms Correlate one half with other half 2 7 2 If X 2 X0 7 Z X I 2 X0 7 Z o H KISS Attitudes Test KAT Please carefully read the statements bellow and indicate the degree to which you agree with each statement using the following scale 1 Strongly disagree 2 disagree 3 Neither Agree nor Disagree 4 Agree 5 Strongly Agree Fire Breathing is the quintessence of cool Men in makeup make me squeal with delight I often wish I was evil incarnatequot Long tongues make me squeal with delight Awwe B Estimating Reliability 2 Internal Consistency Item Covariation a SplitHalf Reliability Problems Assumes that all items are of equal dif culty for performance tests reliability for same data can be different depending on how you half the test Underestimates reliability estimate base on only 12 of actual test length Longer tests are typically more reliable Dilutes the impact of bad items B Estimating Reliability 2 Internal Consistency Item Covariation a SplitHalf Reliability SpearmanBrown Prophesy Formula estimates the reliability for whole test based on the correlation for 12 the test 2r r 0e xx39 1 roe Where rDe correlation between odd and even items Where rxx estimated reliability for full test B Estimating Reliability 2 Internal Consistency Item Covariation a SplitHalf Reliability Spearman Brown Prophesy Formula General Form estimates reliability r XXy of a test that is n times longer I ll3908 2 1 r0en 1 rxx 39 B Estimating Reliability 2 Internal Consistency Item Covariation a SplitHalf Reliability Spearman Brown Prophesy Formula Special Case Estimates the number of times longer n a test will need to be to reach a desired level of reliability rxxy rxx391 roe roe1 roe B Estimating Reliability 2 Internal Consistency Item Covariation b Chronback s Alpha Average intercorrelation between all of the items in a scale All items should be responded to consistently Minimum of 70 Represents all possible split half reliabilities B Estimating Reliability 2 Internal Consistency ltem Covariation b Cronbach s Alpha Item Intercorrelations x Fawn Z 1 f lmX Xl 2X7XZ KAT Example rl2vr13vr14vr23vr24vr34 B Estimating Reliability 2 Internal Consistency Item Covariation b Cronbach s Alpha Conceptually Er 0 Z quotIf Z Mitzi 2 More Precisely K07 a 2 1 K 1FU B Estimating Reliability 3 a Internal Consistency Error Estimation KuderRichardson Formula 21 Used for performance tests correctincorrect item scoring Very useful when item level data is not available Assumes all items are of equal difficulty Tends to be a conservative estimate of reliability B Estimating Reliability 3 Internal Consistency Error Estimation a Kuder Richardson Formula 21 K K 2 K 1 K0392 Where K of items 02 variance of scores 02 21XZ gEX lZN N K1221 KISS Information Standard Skill F am KISSFl Questions Items 1 What band did Gene Simons and Paul Stanley quit to form KISS 2 In what state did KISS start their career 3 What Instrument does Ace Frehley Play 4 Which Original Band Member wore cat face makeup 5 On what album lid KISS rst appear without makeup 7 6 Which band member appeared on National Public Radio and claimed to have slept With over 5000 women7 Answers 1 Wicked Lester 2 New York 3 Lead Guitar 4 Peter Criss later replaced by Eric Car died Nov 1991 and replaced by Eric Singer 5 Lick It Up 6 Gene Simmons
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'