Mod 5 - Graphing homework

Smile Assignment Week 4

Martha Ramsey

Saint Leo University

Research Method II: PSY 535

Instructor Andrea Goldstein

November 20, 2022

Smile Assignment Week 4

The table below shows descriptive statistics for the #correct male and #correct female. The SD for males is 2.769 with a mean of 9.29, while the SD for females is 2.471 with a mean of 5.76. The two statistics are very low, hence affirming the low expectation.

The table below shows the test of the assumption of homogeneity of covariance. This test is offered when “Homogeneity tests” in the options menu are selected (Kim & Cribbie, 2018). The Box's M Test is statistically significant, F (3, 818359.675) = 3.247, p = .021. This statistic shows that the assumption of covariance homogeneity has not been met.

The table below shows the multivariate tests of the effects of “gender” and the interaction between “gender” and “your gender.” The multivariate test can be used instead of univariate tests because they do not require sphericity assumption (Adam & Mujib, 2020). From the table, it is seen that the effect has lower satisfaction power.

The table below provides the test of sphericity. As you can see, the assumption of sphericity has NOT been met, X2 (2) = .000, p = .000.

The next table provides the univariate tests of the within-subjects variable, "gender," and the interaction between the within-subjects and between-subjects variables, "gender" and "your gender." The test of the "gender" variable determines if there are differences between the means at each measurement time, ignoring the between-subjects variable, "your gender." Here, it is statistically significant, F (1, 47) = 38.183, p < .001. The test of the interaction, “gender*your gender," determines whether groups differ in how much they change over time. In other words, do the treatment and control groups change at different rates? We care most about this test. The test is not statistically significant here, indicating that the two groups did not change at different rates, F (2, 16) =. 3.820, p = .057. In other words, the individual’s gender was ineffective in determining a smile.

The table below provides within-subjects contrasts. SPSS uses "polynomial" contrasts as the default option. These can determine if the trend line is linear or curvilinear. Here, the results indicate that the trend line is linear.

Levene’s test in the following table tests the assumption of homogeneity of variance. This is tested at each time period. The assumption has been met at all 2 time periods.

The table below provides the test of between-subjects effects. This is a test of the difference between the groups, ignoring the “gender” variable (Jonsson et al., 2021). This is not usually a critical test, but it may be of interest in certain circumstances. Here, it is not statistically significant, F (1, 47) = .785, p = .785.

References

Adam, M., & Mujib, U. (2020). Students critical-creative thinking skill: A multivariate analysis of experiments and gender.  International Journal of Cognitive Research in Science, Engineering and Education,  8(S), 49-58. https://cyberleninka.ru/article/n/students-critical-creative-thinking-skill-a-multivariate-analysis-of-experiments-and-gender

Jonsson, B., Wiklund-Hörnqvist, C., Stenlund, T., Andersson, M., & Nyberg, L. (2021). A learning method for all: The testing effect is independent of cognitive ability.  Journal of Educational Psychology,  113(5), 972. https://psycnet.apa.org/psycarticles/2020-72600-001.pdf

Kim, Y. J., & Cribbin, R. A. (2018). ANOVA and the variance homogeneity assumption: Exploring a better gatekeeper.  British Journal of Mathematical and Statistical Psychology,  71(1), 1-12. https://bpspsychub.onlinelibrary.wiley.com/doi/abs/10.1111/bmsp.12103