SPSS Flashcards

1
Q

What are the six steps in typical psychometric data analysis?

A

Check data entry is correct - out of range values?
Score the measure (and any others required)
Check the distribution - normal/skewed?
If not normal, non-linear transformation (sqrt, log10?)
Calculate reliability (Cronbach’s, KR-20)
Run analyses (correlations, t-tests) to test your validity hypotheses

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

How do you check data entry/out of range values in SPSS? (x3)

A

Select: AnalyseData - DescriptiveStatistics - Frequencies
Highlight items of interest, paste into ‘variables’ box
Eyeball frequency table/histogram - restriction of range? (if so, crappy discrimination)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

If examination of frequency of responses in SPSS shows restriction of range, what might be the solution? (x1)

A

Reword until you get a more normal distribution

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What is involved in ‘scoring the measure’ in SPSS? (x2)

A

Recode any reversed variables

For questionnaires - work out a total (or mean) for each Ps

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What are the steps to recode reverse items in SPSS? (x5)

A

Select: Transform - RecodeIntoDifferentVariables,
Select variables to be reversed & paste into middle box
Choose a name for new variable
Click ‘old and new values’ - type in (e.g. 5 - 1, 4 - 2…)
Click OK

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

How do you calculate the total or mean score for each Ps in SPSS? (x5)

A

Select: Transform - Compute
Choose name for target variable
Choose function from drop-down
Paste in items of interest from list of variables in left box
Hit OK - new variable will appear at far right of data view

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

How do you check the distribution for skew in SPSS? (x2)
What is an acceptable range for skew in this course?
Why don’t we defer to the criterion used by SPSS? (x1)

A

Select: Descriptives - Options - tick Skew (to calculate the statistic)
Graphs - LegacyDialogs - Histogram (to view graph)
-1 to +1
It’s arguably only appropriate for small samples (

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

How do you do a non-linear transformation in SPSS (ie correct skew)? (x5)

A
Select: Transform - Compute
Choose name for target variable
Chose function from drop-down (log10, sqrt)
Paste item between brackets
Hit OK
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What can you do if the log10 transformation gives you an error message due to Ps with zero scores? (x1)

A

Recode the variable to start from 1

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

How do you calculate Cronbach’s alpha in SPSS? (x3)

A

Select: Analyse - Scale - ReliabilityAnalysis
Paste your items into the ‘items’ box
Hit OK

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

How do you run a correlation in SPSS? (x4)

A

Select: Analyse - Correlate - Bivariate
Paste your score variable (sum or mean) into the variables box
Select Spearman if doing non-parametric (Pearson is auto set)
Click OK

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

How do you do a t-test in SPSS? (x5)

A

Select: Analyse - CompareMeans - IndependentSamplesT-Test
Paste the dependent variable (test sum/mean) into Test Variables box
Paste IV into Grouping Variable
Define Groups - code the two levels (e.g. male = 1, female = 2)
Click Continue, then OK

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

When interpreting SPSS t-test results, what should you do if the Levene’s Test for Equality of Variance is greater than .05? (x1)
Why? (x1)

A

Take your results from the bottom row,

The variances of your 2 groups are significantly different

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What is the non-parametric equivalent of an independent samples t-test? (x1)
How would you run this in SPSS? (x2)

A

Mann-Whitney U
Select: Analyse - NonparametricTests - LegacyDialogues
Choose ‘2 independent samples’

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What is the non-parametric equivalent of the paired t-test? (x2)
How would you run this in SPSS? (x2)

A

Wilcoxon test or Sign test
Select: Analyse - NonparametricTests - LegacyDialogues
Choose ‘2 related samples’

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What is the non-parametric equivalent of the one way unrelated ANOVA? (x1)
How would you run this in SPSS? (x2)

A

Kruskal-Wallis
Select: Analyse - NonparametricTests - LegacyDialogues
Choose ‘K independent samples’

17
Q

What is the non-parametric equivalent of the one way repeated measures ANOVA? (x1)
How would you run this in SPSS? (x2)

A

Friedman
Select: Analyse - NonparametricTests - LegacyDialogues
Choose ‘K related samples’