Sarah Hobson

BUSI 820: Quantitative Research Methods

This study source was downloaded by 100000838887847 from CourseHero.com on 12-17-2021 05:23:01 GMT -06:00

https://www.coursehero.com/file/109502641/BUSI-820-Week-8-DBFdocx/
D.8.9.6 In Output 9.6: (a) Describe the F, df, and p values for each dependent variable as

you would in an article. (b) Describe the results in nontechnical terms for visualization and

grades. Use the group means in your description.

(a) The One-Way ANOVA test was used to check if there are statically significant

differences between the father’s education revised across the grades in high school, visualization

test scores, and math achievement. The ANOVA table showed that there were statistically

significant differences in the father’s education groups on grades in high school is F (2, 70) =

4.091, p =0.021. (b) There was a statistically significant difference between father’s education

revised groups on math achievement showing F (2,70) = 7.881, p = 0.001. There were not any

statistically significant differences between the father’s education revised groups and student’s

visualization scores, which showed F (2, 70) = 0.763, p = 0.470. In the Test of Homogeneity of

Variances table, the data provides Levene’s test to check the assumption that the variances of the

three father’s education groups are equal for each of the dependent variables. For ANOVA, the

researcher is interested in the test based on means. For grades in h.s. (p = .220) and visualization

test (p = .153) the Levene’s test are not significant. Thus, the assumption is not violated.

However, for math achievement, p = .049; therefore, the Levene’s test is significant and the

assumption of equal variances is violated. In this latter significance, the researcher could use the

similar nonparametric test (Kruskal-Wallis). Or, if the overall F is significant, the researcher

could use a post hoc test designed for situations in which the variances are unequal (Morgan et

al., 2020).

D.8.9.7 In Outputs 9.7 a and b, what pairs of means were significantly different?
In Output 9.7, only the means cores for high school graduate or less and BS are more

were significantly different; the other two comparisons are not significantly different. For 9.7b,

the mean scores for high school graduate or less and BS or more was significantly different, and

This study source was downloaded by 100000838887847 from CourseHero.com on 12-17-2021 05:23:01 GMT -06:00

https://www.coursehero.com/file/109502641/BUSI-820-Week-8-DBFdocx/
also the difference between high school graduate or less and some college was significant

(Morgan et al., 2020).

D.8.9.8 In Output 9.8, interpret the meaning of the sig. values for math achievement and

competence. What would you conclude, based on this information, about differences

between groups on each of these variables?

A Kruskal-Wallis nonparametric test was conducted to test for statistically significant

differences between father’s education groups in math achievement because there were unequal

variances and ns across groups. The test indicated that the three father’s education groups

differed on math achievement, x2 (2, N = 73) = 13.38, p = .001. Post hoc Mann-Whitney tests

compared the three pairs of father’s education groups on math achievement, using a Bonferroni

corrected p value of .017 to indicate statistical significance. The mean rank for math achievement

of students whose fathers had some college (36.59, n = 16) was significantly higher than that of

students whose fathers were high school graduates or less (23.67, n = 38), z = 2.76, p = .006, r = .

38, a medium to large effect size (Cohen,1998; Ostertagova et al., 2014). Also, the mean rank for

math achievement of students whose fathers had a bachelor’s degree or more (38.47, n = 19),

was higher than that of students whose fathers were high school graduates or less (24.26, n =38),

z = 3.05, p = .002, r = 40, a statistically significant difference with a medium to large effect size.

There was not a statistically significant difference in math achievement between students whose

fathers had some college and those whose fathers had a bachelor’s degree or more, z = -1.23, p

= .23, r = .21. The effect size for this comparison was small to medium, suggesting that it could

be worthwhile to replicate this study with a larger sample (Morgan et al., 2020).

D.8.9.9 Compare Outputs 9.6 and 9.8 with regard to math achievement. What are the most

important differences and similarities?

This study source was downloaded by 100000838887847 from CourseHero.com on 12-17-2021 05:23:01 GMT -06:00

https://www.coursehero.com/file/109502641/BUSI-820-Week-8-DBFdocx/
Output 9.6 also provides descriptive statistics of mean math achievement test scores,

while Output 9.8 provides a mean rank for each of the three father’s education groups. Output

9.6 also provides a test of homogeneity of variance to test if the variances of the three groups are

significantly different. In Output 9.8, it is not necessary to have such a test of homogeneity of

variances because the Kruskal-Wallis test is nonparametric and does not have that assumption.

Output 9.6 uses on-way ANOVA to compare statistically the means of the three fathers’

education groups on math achievement test scores. Output 9.8 uses the K-W nonparametric test

to make a similar comparison. The p value is the same (.001) for both outputs, indicating a

significant difference somewhere between group scores. Neither test tells us which pairs of

groups are different (Morgan et al., 2020; Ostertagova et al., 2014).

D.8.9.10 In Output 9.9: (a) Is the interaction significant? (b) Examine the profile plot of the

cell means that illustrates the interaction. Describe it in words. (c) Is the main effect of

academic track significant? Interpret the eta squared. (d) How about the “effect” of math

grades? (e) Why did we put the word effect in quotes? (f) Under what conditions would

focusing on the main effects be misleading?

(a) Usually, the researcher would ignore the lines in the table labeled “corrected model”

(which just summarizes all “effects” taken together) and intercept (which is needed to fit the best

fit regression line to the data) and skip down to the interaction F (mathgr * academic track)

which, in this case, is not statistically significant, F (1, 71) = .337, p = .563. (b) If the interaction

is statistically significant, then the researcher should also analyze the differences between cell

means (the simple effects). The profile plots may be helpful in visualizing the interaction, but the

researcher should not discuss statistically nonsignificant differences because the plots may be

misleading. The profile plots of cell means (which follow the table of between-subjects effects)

This study source was downloaded by 100000838887847 from CourseHero.com on 12-17-2021 05:23:01 GMT -06:00

https://www.coursehero.com/file/109502641/BUSI-820-Week-8-DBFdocx/
assists the researcher in visualizing the nature of a significant interaction when one exists. When

the lines on the profile plots are parallel, there is not a significant interaction. Note that the

researcher requested that the separate lines represent the two academic tracks because they felt

that this would make a significant interaction easier to interpret than if the two lines represented

predominant level of grades. However, really, either independent variable could have been

represented in either part of the graph (Morgan et al., 2020). (c) When it comes to examining the

main effects of math grades and of academic track, both are statistically significant. The

significant F for math grades means that students with fewer As and Bs in math scored lower on

math achievement than those with high math grades (M = 10.81 vs. 15.05), and this difference is

statistically significant (p < .001). Academic track is also statistically significant (p < .001).

Because the interaction is not statistically significant, the “effect” of math grades on math

achievement is about the same for both academic tracks. If the interaction were statistically

significant, the researcher would state that the “effect” of math grades depended on which

academic track the researcher was considering. For example, it might be large for the fast track

and small for the regular track. In addition to this, the callout boxes about the adjusted R squared

and partial eta squared are important. Eta, the correlation ratio, is used when the independent

variable is nominal and the dependent variable (math achievement in this case) is normal

(approximately normally distributed). Eta squared is an indicator of the proportion of variance

that is due to between-groups differences. Because eta and R, like r, are indexes of association,

they can be used to interpret the effect size (Morgan et al., 2020). Eta for academic track is about

.40, which is a medium to large effect size. The overall adjusted R is about .46, a large effect

size. (d) In this example, eta (not squared) for math grades is about .41 and thus a medium to

large effect. Notice that the adjusted R² is lower than the unadjusted (.22 versus .25). The reasons

This study source was downloaded by 100000838887847 from CourseHero.com on 12-17-2021 05:23:01 GMT -06:00

https://www.coursehero.com/file/109502641/BUSI-820-Week-8-DBFdocx/
for this is that the adjusted R² takes into account (and adjusts for) the fact that not just one

variable but three (math grades, academic track, and the interaction) were used to predict math

achievement. (e) The ANOVA table, called Tests of Between -Subjects Effects, is the key table.

Note that the word “effect” in the title of the table can be misleading because the study was not a

randomized experiment. Therefore, the researcher cannot state in the report that the differences

in the dependent variable were caused by or were the effect of the independent variable (Morgan

et al., 2020). (f) If the interaction were significant, the researcher would need to be cautious

about the interpretation of the main effects because they could be misleading (Morgan et al..,

2020).

This study source was downloaded by 100000838887847 from CourseHero.com on 12-17-2021 05:23:01 GMT -06:00

https://www.coursehero.com/file/109502641/BUSI-820-Week-8-DBFdocx/
References
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ:

Erlbaum.

Morgan, G., Leech, N., Gloeckner, G., Barrett, K. (2020). IBM SPSS for Introductory Statistics

(6th Ed.). Routledge.

Ostertagova, E., Ostertag, O., & Kováč, J. (2014). Methodology and application of the Kruskal-

Wallis test. In Applied Mechanics and Materials (Volume 611, pp. 115-120). Trans Tech

Publications Ltd.

This study source was downloaded by 100000838887847 from CourseHero.com on 12-17-2021 05:23:01 GMT -06:00

https://www.coursehero.com/file/109502641/BUSI-820-Week-8-DBFdocx/
Powered by TCPDF (www.tcpdf.org)