LAB INSTRUCTIONS

Assignment Goals:

• Familiarize yourself with independent sample t-test.
• Submit SPSS output and syntax files and a basic apa-style report (1-2 pages)

Dataset:

• 2014 Canadian Community Health Survey (n = 1000).

Research question:

• Do males and females differ in their levels of depression?

Variables of interest:

• Sex (between subject factor/independent variable)
• Depression scale - short form score (dependent variable/outcome variable)

Task:

1. In SPSS, conduct an independent samples t-test to determine whether males and females significantly differ in their depressed mood
2. Write up the results in Word doc

Task 1: SPSS independent samples t-test

1. Conduct relevant assumptions testing
   • AnalyseDescriptivesExplore

To test the normality assumption:

• Move "sex" into the "factor list" box
• Move "depression scale" into the "dependent list" box
• Click "Statistics"
  • Ensure 'descriptives' is checked at 95% confidence intervals. Click 'continue'
  • Click "Plots"
    • Uncheck "stem and leaf"
    • Check 'histogram' and 'normality plots with tests'
    • Leave everything else as "none". Click 'continue'
  • Click 'Paste', then highlight the syntax and execute.

1. Conduct the independent samples t-test

*Remember, the homogeneity of variance assumption testing is included within the t-test*

• AnalyseCompare MeansIndependent-Samples T Test
• In the dialogue box:
  • Put your dependent variable (depression scale) in the "test variable(s)" box
  • Put your independent variable (sex) in the "Grouping variable" box.
    • Click on "define groups"
    • Under "use specified groups": Our groups are coded as "1" (male) and "2" (female), so put these values in the group 1 and 2 boxes. Click 'continue'
    • Click "Options", in the dialogue box, ensure that under 'confidence interval', it says "95%" and "exclude cases analysis by analysis" is selected. Click 'continue'
    • Click "Paste". Then highlight the syntax and execute.

Task 2: Write up the results

• 1-2 pages, apa style, double-spaced
• Include the following:
  • Sample sizes, means, and standard deviations for the two groups
  • Results from the assumptions testing
    • Normality - e.g., Shapiro-Wilk test results with p-value, skewness, kurtosis, visual speculation
    • Levene's Homogeneity of Variance test results
    • Do not just report the numbers. Interpret: Tell us what the numbers mean (e.g., Is the normality assumption violated/met?)
    • Remember, there are 2 additional assumptions that are not subject to SPSS testing - namely, the assumption of independence of observations and whether the dependent variable is interval/ratio. Be sure to comment on whether these assumptions were met.
  • Results of the t-test
    • T-test value, degrees of freedom, p-value, mean difference, confidence intervals for the mean difference
    • Again, do not just report the numbers. Tell us what they mean

(e.g., Is the difference between groups significant? And in what direction - that is, which group was higher/lower on the depression scores?)

Example of APA-style result write-up

A significant difference was found between group 1 and group 2 in the level of anxiety, t(19) = 5.34, p < .05. On average, group 1 had a higher level of anxiety (M = 10.26, SD = 2.47, 95% CI [5.62, 18.31]) than group 2 (M = 3.15, SD = 1.02, 95% CI [-2.62, 8.36]).

What do the notations and numbers mean?

M = mean

SD = standard deviation

95% CI = confidence intervals, such that they are written as [lower bound, higher bound]

t (degrees of freedom) = t-value

p = probability of rejecting the null hypothesis

Submit on blackboard:

• SPSS syntax
  • lastname.firstname.Lab2AssignmentSyntax (your syntax file)
• SPSS output
  • lastname.firstname.Lab2AssignmentOutput (your output file)
• Word write-up
  • lastname.firstname.Lab2AssignmentWriteup (your word file)

RUBRIC For LAB

2pts

• for reporting sample size, means, and standard deviations for each of the two groups (male and female)

4 pts

• commenting on independence of observation; for commenting on whether DV (depression score) is at least interval scale
  • homogeneity of variance assumption ( Levene's test statistic, p-value, and interpretation)
  • normality assumption Shapiro-Wilk test statistic, p-value, degrees of freedom, interpretation (normality assumed/violated?)
  • skewness & kurtosis values, within cutoffs or not, interpretation of it (normality assumed/violated?)
  • Histogram (present and mentioned in text)

9 pts

• Correct form of reporting of t-test
• interpreting significance
• direction of significance (i.e., which group was higher/lower on depression scores)
• mean difference & confidence intervals for the mean difference

• APA and overall writing
• including both syntax and output files