8210 WK 6 DISCUSSION

3

Carey-Ann Thurlow 

RE: Discussion - Week 6

COLLAPSE

Top of Form

Research Design and t-Tests

Within Examining the relationship between student test anxiety and webcam-based exam proctoring, Kolski & Weible (2018) undertook an exploratory correlation research design to determine common behaviors in higher education students. The correlation research design measured test anxiety variables experienced by students in settings where online exam proctoring was utilized, while also examining whether their behaviors were associated with the online environment (Walden University, 2022). Although it is difficult to determine cause and effect in a correlation design, it is useful in this study because correlation is necessary to demonstrate causation (Walden University, 2022).

Kolski & Weible (2018) chose this research design in their exploratory, mixed methods study to determine a correlation between test anxiety and behaviors in online learning. Participants were e-learning students from a large 4-year university and a 2-year community college, who completed four to five exams with recorded footage for analysis, along with interviews (Kolski & Weible, 2018). The research consisted of a self-report measure of anxiety, using a Test Anxiety Inventory (TAI), consisting of 20 questions and a four-point Likert scale.

Using descriptive statistics, parametric inferential statistics and correlation procedures, Kolski & Weible (2018) analyzed the observed behaviors, exam scores and TAI scores. A Pearson correlation coefficient was used to examine relationships between TAI scores and the top ten most frequently observed behaviors during online examinations (Kolski & Weible, 2018). The researchers also used a one-way Anova test comparing TAI scores and twenty-three total observed behaviors (Kolski & Weible, 2018). While eyebrow furrowing, lip biting and clearing of the throat were analyzed in the final Pearson correlation, Kolski & Weible (2018) found no other behaviours to have a significant correlation (p<.05).

Kolski & Weible (2018) chose the most appropriate design and analysis for their research, as they collected information on a sample population without manipulating the variables, unlike that of an independent t-test. The data was displayed in tables and graphs, displaying the top ten behaviors associated with online proctoring (smallest to largest), in addition to the top twenty-three behaviors. The results were appropriately displayed with explanatory tables, followed by a detailed explanation of the data. Kolski & Weible (2018) reported the effect size as over 200 participants, which is meaningful due to the sample being N>50. Out of the 272 participants, 238 chose online proctoring, which allowed for a robust account of recorded behaviors during examinations. This research presents an alternative view surrounding the benefits of online exam proctoring, along with student comfort in an online testing environment.

 

References

Kolski, T., & Weible, J., (2018). Examining the relationship between student test anxiety and webcam-based exam proctoring. Online Journal of Distance Learning Administration, 21(3), 1–15.

Walden University. (2022). Skill builders. Research design and statistical design.  https://content.waldenu.edu/d1c00f22444bfb7cf79c9487accceada.html

Bottom of Form

Letitia Melvin 

RE: Discussion - Week 6

COLLAPSE

Top of Form

Which is the research design used by the authors?

 

The following article demonstrates how to choose the appropriate test to collect, analyze, and interpret data to report research findings appropriately. Montañez et al. (2015) examined the effect of Turn 2 Us (T2 Us), a school-based mental health promotion and prevention program (SBMH-PP), on predominately Latino at-risk students at two urban elementary schools. A quasi-experimental research design was utilized a questionnaire and survey (pre and post-intervention). During the 2088-2009 school year, a sample size of 174 students made up of 62% fifth-graders, 30% fourth grades, and 8% third graders participated in the Turn 2 Us program. The student demographic was 87% Latino, 10% African American, and 3 % other (51% male, 49% female). A quasi-experimental research design was used, which is appropriate because the authors used and an existing class of students from two elementary schools.

 

Why did the authors use this t test?

Do you think it’s the most appropriate choice? Why or why not?

 

T test is a statistical test that compares the means of two samples. According to Montañez et al. (2015), paired-sample t tests were performed to compare T2 U students’ Math and ELA scores and absences. Paired sample t tests compare a sample at two different times (Frankfort-Nachmias et al., 2020). The authors compared the Math and ELA scores and absences from the previous year to the year during the program's participation, which indicates a comparison was made at different times. It would have been appropriate for the authors to use the paired-samples t test if they were comparing two variables, but they were comparing Math, ELA, and absences. The authors could have used a three-way ANOVA to compare Math and ELA scores and absences. ANOVA is a statistical technique that allows the examination of multiple variables (Frankfort-Nachmias et al., 2020). An independent sample t test or two-sample t test was used to compare absences between T2 U participants and the control group of non-T2 U students (Montañez et al., 2015). An independent samples t test compares the means of two independent samples (Frankfort-Nachmias et al., 2020). The independent sample t test was appropriate because the sample size was larger than 50 (N=174), and the comparison of absences between two samples (T2 U and non-T2 U students). The authors randomly selected a sample and divided them into groups in which T2 U students are independent of non-T2 U students.

 

Did the authors display the data?

 

Montañez et al. (2015) used descriptive tables and a line graph to display the data. Table 1 shows the results of the Strength and Difficulties Questionnaire (SDQ), which was used as a baseline measure of student risk for mental health problems after being enrolled in the program. Figure 1 provides the report from the Student Assessment Survey (SAS), pre- and post-intervention track, completed by the teachers. Lastly, the authors displayed in Table 2 the means and standard deviations of the students’ academic achievement (Math and ELA) and absences.

 

Do the results stand alone? Why or why not?

 

In Table 1, the authors displayed the SDQ five subscales (25 items) in a descriptive table which showed that 32 % (N=174) of the students showed a high risk for mental health issues. It does not reveal the gender or grade of each school. It summarizes both of the elementary schools that participated in the survey. Figure 1 shows the SAS pre- and post-intervention results in a line graph that may not have been the best, but it is does give a visual of the results. Table 2 shows the Math and ELA means and standard deviation test scores along with attendance for the previous year and the year the students were in the program. Overall, the three visual display results can stand alone.

 

Did the authors report effect size? If yes, is this meaningful? 

 

The authors reported the effect size using 0.01=small effect, 0.06=moderate effect, and 0.14=large effect. The Strength and Difficulties Questionnaire (SDQ) had a moderate effect size (0.07), and the Student Assessment Survey (SAS) had a low (0.050 and large (0.39) effect size. The ELA scores indicated a large effect size (0.27), while Math scores (0.10) and absences (0.06) indicated moderate effect sizes. Montañez et al. (2015) reported the data as being statistically significant and meaningful. The effect sizes indicated that students from low-income communities at high risk for mental and behavioral challenges benefit mentally and academically from a school-based mental health and behavior program.

 

 

 

 

 

 

References

 

Frankfort-Nachmias, C., Leon-Guerrero, A., & Davis, G. (2020). Social statistics for a diverse society (9th ed.). Thousand Oaks, CA: Sage Publications.

 

Montañez, E., Berger-Jenkins, E., Rodriguez, J., McCord, M., & Meyer, D. (2015). Turn 2 Us: Outcomes of an Urban Elementary School-based Mental Health Promotion and Prevention Program Serving Ethnic Minority Youths. Children & Schools, 37(2), 100–107. https://doi.org/10.1093/cs/cdv004

Bottom of Form