Points: 50
Statistical tests are used to conduct hypothesis testing, requiring one to have a fully articulated and clear research question and null hypothesis. For testing for group differences, the following model for a research question and hypotheses should be used:
RQ: To what extent, if any, is there a difference of <dependent variable> between <independent variable> groups?
H0: There is not a significant difference of <dependent variable> between <independent variable> groups.
H1: There is a significant difference of <dependent variable> between <independent variable> groups.
Recall the independent variable must be continuous (quantitative) with an interval or ratio level of measurement and the dependent variable must be categorical and for independent-samples t-tests, there may be only two groups. The example used in this post will test the following null hypothesis:
RQ: To what extent, if any, is there a difference of Neuroticism between gender groups?
H0: There is not a significant difference of Neuroticism between gender groups.
H1: There is a significant difference of Neuroticism between gender groups.
Where the dependent variable is Neuroticism, as measured by IPIP-50, an interval level of measure, and the independent variable is gender, which has the classes of male and female. An independent-samples t-test will be used to test the null hypothesis.
Prior to conducting an statistical test, one must test for assumptions. To see this completed for the independent-samples t-test, visit the Data Screening: Between Groups Output Assignment.
With the tests of assumptions completed, one completes the independent-samples t-test. See the video below for a walk-through of how to accomplish this test in SPSS.
To write up the results, begin with an introduction to state what was accomplished.
An independent-samples t-test was accomplished to test the null hypothesis of the following:
RQ: To what extent, if any, is there a difference of Neuroticism between gender groups?
H0: There is not a significant difference of Neuroticism between gender groups.
H1: There is a significant difference of Neuroticism between gender groups.
Next, describe the process of the data screening and tests of assumptions. Doing so demonstrates to the reader that the data meet the criteria necessary to complete the tests. This is an important step in completing a statistical analysis. A write up of the tests of assumptions is demonstrated in the Data Screening: Between Groups Output Assignment.
When interpreting the results, be sure you are using the correct data. The table below shows the output from SPSS for the independent-samples t-test. The red box indicates the results of the independent-samples t-test. Note, the columns to the left indicate the results of Levene’s test, which was discussed in the data screening post. In the example, the results of Levene’s test is significant (see the data screening post for an explanation). Because of this, you should interpret the second row of results, which corresponds to the Equal variances not assumed results, as shown by the blue arrow. To report the results, use the format of t(df) = t-test statistic, two-sided p-value. The reported results from Figure 1 are t(1179.256) = -2.153, p = .031.
Figure 1
Interpreting the SPSS Output
Lastly, one presents the results of the independent-samples t-test and interprets the results in terms of the null hypothesis. As a reminder, the results either reject or fail to reject the null hypothesis. The results DO NOT accept any hypothesis and there is no test of the alternative hypothesis, so one should not reference such a test. Also, the researcher does not reject or fail to reject the null hypothesis, either. The researcher’s role is establish the rules under which hypothesis testing will occur, such as predetermining the level of significance to be used. Once the rules are established, the results determine the outcome of the hypothesis testing. Again, the results either reject or fail to reject the null hypothesis.
An independent-samples t-test (equal variances not assumed) demonstrated a statistically significant difference of Neuroticism between males (M = 10.89, SD = 2.83), and females (M = 11.26, SD = 3.30) with results of t(1179.256) = -2.153, p = .031 (see Table 1). The results reject the null hypothesis, which states there is not a significant difference of Neuroticism between genders.
Submitting the assignment
When submitting this assignment, you must first describe the assignment. The write-up should be descriptive of the variables and the activities used to screen the data, along with a description of the results. All submissions must be a single Microsoft Word document. Do not submit the SPSS file.