Skip to Main Content

Statistics Resources

This guide contains all of the ASC's statistics resources. If you do not see a topic, suggest it through the suggestion box on the Statistics home page.

Dependent Samples T-test

The dependent samples t-test is used to compare the sample means from two related groups.  This means that the scores for both groups being compared come from the same people. This test is also referred to as paired samples t-test. The purpose of this test is to determine if there is a significant change from one measurement to the other.

Assumptions

  1. The dependent variable is continuous (interval or ratio) - based on operationalization of the variable
  2. Categorical independent variable with two related groups - based on operationalization of the variable (can be test/retest or matched pairs)
  3. Each observation is independent of other observations - evaluated through the study's design
  4. No significant outliers in the differences between the two scores - can be assessed using boxplots, scatterplots, and other methods
  5. The distribution of the differences should be approximately normally distributed - can be assessed using histograms, Q-Q plots, skewness, kurtosis, K-S test and/or Shapiro Wilks

Running a Single Sample t-test using SPSS

  1. Analyze > Compare Means and Proportions > Paired-Samples T Test
  2. Move the dependent variable data for group 1 into the Variable1 box.
  3. Move the dependent variable data for group 2 into the Variable2 box.
  4. Review Options for missing values and confidence interval options.
  5. Select OK to run the analysis.

Interpreting the Output

  • Paired Samples Statistics
    • provides summary statistics about the dependent variable for each group including sample size, mean, standard deviation, and standard error of the mean
  • Paired Samples Test
    • provides the results of the statistical test
      • test statistic = t
      • associated probability (p-value) = Sig.
      • confidence interval of the difference (if elected to include)

Reporting Results in APA Style

When reporting the results of the dependent samples t-test, APA Style has vague requirements on what information should be included. Below is the key information you should anticipate reporting when presenting the results of the test. You want to replace the red text with the appropriate values from your output.

t(degrees of freedom) = the t statistic, p = p value.

Example:

A dependent-samples t-test was run to determine if long-term recall improved with the introduction of the Say it Again memorization technique. The results showed that the average number of words recalled without this technique (M = 13.5, SD = 2.4) was significantly less than the average number of words recalled with this technique (M = 16.2, SD = 2.7), (t(52) = 4.8, p < .001).

Notes:

  • When reporting the p-value, there are two ways to approach it. One is when the results are not significant. In that case, you want to report the p-value exactly: p = .247. The other is when the results are significant. In this case, you can report the p-value as being less than the level of significance: p < .05. If SPSS reports the p-value as < .001, you would report it the same way: p < .001.
  • The t statistic should be reported to two decimal places with a 0 before the decimal point if needed: t = 0.36 or t = 2.71.
  • Degrees of freedom for this test are n - 1, where "n" represents the number of pairs in the sample. n can be found in the SPSS output.

Was this resource helpful?