Statistical Data Analysis: Research Proposal

Anytime a researcher or psychologist is wanting to begin a project, or research assessment, it is important to understand the methods in which data can be analyzed. According to an article published by James Cook University (2020) there are five main types of research methods. The first is Naturalistic Observation—this is typically when a subject’s environment cannot be recreated in a lab setting, therefore the researcher is forced to go to the subject’s in their natural environment. It is described that a major disadvantage to utilizing this research method is due to the fact that individuals will typically react or respond differently to stimulus when they know that they are being watched or observed. The second method of research is called the Survey method and is typically chosen utilizing the random sample method. However, with the survey method, the data being analyzed is typically respective to an entire population, therefore there is heavy emphasis on the types of questions that should and should not be asked (James Cook University, 2020). The third is Experimental Methods—these are studies that can be conducted while in a lab or can be reproduced in a lab for validation purposes. The fourth is Correlational Studies—this method is most commonly done when an experiment type of method is not possible, and usually focuses on a combination of Naturalistic Observation, Surveys, and Archival Research. James Cook University states that Correlational Studies typically make understanding of variables we cannot control possible, but they do not provide enough information to allow a researcher to formulate a cause and effect from the study. The fifth is Case Study—which is typically focused on a particular individual or particular group over a long period of time, and due to long-term participation required from participants there have been ethical concerns raised about Case Studies (James Cook University, 2020).

After a research method has been decided upon based on the type of information the researcher is looking to obtain, it is then important to remember that the four main objectives of psychological research are (1) to describe, (2) to explain, (3) to predict, and (4) to change or control behaviors (Bourchrika, 2023). These objectives are the most basic foundations to theories and studies that attempt to explain any cognitive, emotional, or behavioral processes that individuals may face in their daily lives (Bourchrika, 2023).

For the purposes of this assessment, this learner has chosen to focus her attention on the grade level of students, compared to if the students passed or failed the class, based on the dataset provided in the course room. Social Learning Theory could be best used to articulate if a student will pass or fail the class, based on social expectations set forth by the school, their parents, or their peers. Self-esteem is a construct that can be applied to many different aspects of life—including, but not limited to “improved physical health, relationship satisfaction, and positive and negative effect,” (UK Essays, 2018). Self-esteem is also an “evaluation of an individual’s beliefs and attitudes toward his or her abilities and values,” (Zhao et al, 2021). When we assess the relationship between self-esteem and Social Learning Theory, the two can seemingly go hand-in-hand.

The four main elements that go into Social Learning Theory is (1) Attention, individuals cannot learn if their attention is not focused on the task at hand, and an individual’s ability to pay attention is largely related to the behavior observed, the complexity, and/or the perceived value; (2) Retention, individuals can internalize information, behaviors and their consequences; (3) Reproduction, individuals can reproduce learned behaviors and receive feedback to adjust or modify their approach for future interactions; and (4) Motivation, an individual’s willingness to perform based on rewards or punishment (Western Governors University, 2023). An individual’s motivation can be directly linked to their self-esteem (Perera, 2022).

Researchers theorize that self-esteem can be an added stressor on a student’s performance in the manner that it can add stress in already stressful environments. A study relayed that individuals with a high level of self-esteem had added levels of test anxiety and more tension on certain tasks (Van Der Kaap-Deeder et al, 2016). Within this study we will be able to analyze, with potentially limited information, the relationship between students’ self-esteem and their motivation within the Social Learning Theory to complete and pass the enrolled course.

The hypothesis for this study will be that the upperclassmen will most likely succeed well in the class, while the underclassmen will be unlikely to succeed well, and any individuals in their Senior year would be expected to do better than the third-year students. The independent variable will be upperclassmen (specifically the juniors and seniors) and lowerclassmen (specifically the freshman and sophomores) grade levels as defined as which grade level the students are enrolled in. The dependent variable will be whether the students pass the class or fail the class. The number of upperclassmen (juniors and seniors) that pass the class compared to the lowerclassmen (freshman and sophomores) will indicate that the upperclassmen students were more prepared for this class. Failing this class is defined as not having the necessary skills to be prepared for the class, whereas passing the class would be defined as having the necessary skills to succeed well in the class—including skills such as stress management, dedication to graduate, and higher self-esteem to perform well under stress. The data is considered as nominal.

Methods

Sampling

I will use the non-probability sampling, which will specifically be the convenience sampling method because the data is archival data, and is not considered randomly selected from the population, and will recruit my sample from grades.xlsx (Capella University, 2023). When utilizing convenience sampling there is essentially “no pattern whatsoever in acquiring these respondents,” (Galloway, 2005) because the data has previously already been acquired, leaving little room for myself to adjust any data collected.

Measures

I will use hypothesis testing to measure and test assumptions and draw conclusions based on the data provided from the population to find the confidence interval on whether upperclassmen (juniors and seniors) perform better in the class compared to underclassmen (freshman and sophomores). I will be utilizing the quasi-experimental sampling method because the data is archival data, and will recruit my sample as described above. The advantages to using a quasi-experimental designed method is that you can, “mimic an experiment and provide a high level of evidence without randomization,” (Office for Health Improvement and Disparities, 2021). A couple of drawbacks to utilizing this method is additionally that a researcher would not be able to “rule out other factors out of your control” that caused the results of the data, and it can be difficult to choose an appropriate comparison group, (Office for Health Improvement and Disparities, 2021).

Procedure

I will use a chi-square test to determine how significant the differences between the students who passed, and the students that failed are. This is considered the best method for this type of research due to the fact that a chi-square statistic will be able to determine results that categorical, and evaluate the tests of independence in each category (i.e., upperclassmen and passing or failing the class) (Quantitative Results Section, n.d.). All participants will be volunteers. The participants will complete a demographic questionnaire to get information on gender, ethnicity, quiz scores, final scores, and GPA. All participants will be in one group. The participants will report their grade level. Lastly, participants will report whether they passed or failed the class. Potential variables affecting this study may include things such as gender or ethnicity, and this learner will review these factors to determine if they are a factor or not. Differences between individuals that may have an effect are the section in the class that they are in, whether they sit closer to the front of the classroom or in the back of the classroom, potentially receiving more distractions. However, being that this data has previously been recorded and acquired, and is now considered as archival data, there is an ability to control any of these additional facts. Therefore, while they can be noted, they will not be studied or assessed in great length during the course of this study.

Demographic Statistics

The archival database used for this study contained demographic variables describing the sample including identified gender, ethnicity, and year in school. The majority of the sample identified as female (n = 58, 55.2 %) while a minority identified as male (n = 37, 35.2%), transgender (n = 3, 2.8%), non-binary (n = 4, 3.8%), and those that did not wish to disclose their gender identity (n = 3, 2.8%). The majority of the sample additionally identified as white (n = 45, 42.8%), black (n = 24, 22.8%), Asian (n = 20, 19%), Hispanic (n = 11, 10.4%), and native American (n = 5, 4.7%). The majority of participants juniors, (n = 64, 60.95%), sophomore (n = 19, 18.09%), seniors (n = 19, 18.09%), freshman (n = 3, 2.8%).

Statistical Measures Used for this Study

This study selected students who were in the upperclassmen (juniors and seniors) and lowerclassmen (freshman and sophmores) levels to provide group comparisons of their rate of passing or failing the selected class. Chi-square test was utilized to examine the hypothesis. A chi-square test is described as a statistical test “used to compared observed results with expected results,” (University of Southampton, n.d.). Therefore, Chi-square test is method selected to help better understand and interpret the relationship between class levels and the pass/fail rate.

Statistical Findings

The number of students who passed in the upperclassmen group was (n = 78, 93.9%) and the number of students who passed in the lowerclassmen group was (n = 22, 100%). The percentage of students who passed in the lowerclassmen group was higher than those in the upperclassmen group. To determine if this difference was significant or not, a chi-square test analysis was performed. The findings of this analysis demonstrated the relationship between these variables was not significant, X2 (1, N = 2.657) = 3, p = .448. The effect sizes with Cohen’s d=(1 – 1)/1.135187 = 0 when comparing Sophomores to Seniors, lowerclassmen to upper class students that had the biggest statistical differences; Glass’s delta = (1 – 1)/0.905 = 0 and Hedges’ g = (1 – 1)/1.250846 = 0 (Social Science Statistics, n.d.).

Interpretation of Findings

Based on the information provided for the use of this assessment, I can conclude that the lowerclassmen appears to have preformed better in this class, compared to the upper-class students. However, based on the population sizes (21 lowerclassmen and 74 upperclassmen, with only 5 upperclassmen failing the class) it would appear that given the population size of the upperclassmen the percentage of upperclassmen compared to the whole class that passed (68%) is still greater than the population size of the lowerclassmen that passed (20%) and with a p value of 0.448 the results are not significant.

I can additionally conclude that this class was most likely aimed toward a third-year student, a Junior, and that the majority of Juniors that took this course did in fact pass the class. A factor to consider when analyzing the data is that upper class students are typically under more stress than the lowerclassmen as well, as they are preparing for further college degrees or careers or additional concerns such as stress management, dedication to graduate, and higher self-esteem to perform well under stress.

References

Bragg, E. W. (1968). A Study of the College Campus as a Learning Environment. Tennessee A

and I State University. https://www.jstor.org/stable/2294083

Bourchrika, I. (2023). The Four Goals of Psychology. Research.Com.

https://research.com/education/goals-of-psychology#:~:text=To%20sum%20up%

2C%20the%20four,face%20in%20their%20daily%20lives.

Capella University. (2023). Grades.xlsx. Capella University Courseroom.

Cuemath. (2023). Inferential Statistics. Cuemath. https://www.cuemath.com/data/inferential-

statistics/

Glen, S. (n.d.) T Test (Student’s T-Test): Definition and Examples. Statistics How To.

https://www.statisticshowto.com/probability-and-statistics/t-test/

Hanson, M. (2022). College Dropout Rates. Education Data Initiative.

https://educationdata.org/college-dropout-rates

James Cook University. (2020) Five Types of Psychological Research. James Cook

University Australia. https://online.jcu.edu.au/blog/five-types-of-psychological-research-you-should-know

Office for Health Improvement and Disparities. (2021). Quasi-Experimental Study: Comparative

Studies. Gov.Uk. https://www.gov.uk/guidance/quasi-experimental-study-comparative-studies

Perera, K. (2022). Self Esteem and Motivation (Why is it so important?). More Self Esteem.

https://more-selfesteem.com/self-motivation-and-self-esteem/

Quantitative Results Section. n.d. Using Chi-Square Statistic in Research. Complete

Dissertation by Statistics Solutions. https://www.statisticssolutions.com/free-resources/directory-of-statistical-analyses/using-chi-square-statistic-in-research/

Social Science Statistics. n.d. Effect Size Calculator for T-Test. Social Science Statistics.

https://www.socscistatistics.com/effectsize/default3.aspx

UK Essays. (2018). Social Learning Theory Self Esteem Psychology Essay. UK Essays.

https://us.ukessays.com/essays/psychology/social-learning-theory-self-esteem-psychology-essay.php

University of Southampton. (n.d.) Chi-Square. University of Southampton.

https://www.southampton.ac.uk/passs/full_time_education/bivariate_analysis/chi_square.page#:~:text=A%20chi%2Dsquare%20test%20is,the%20variables%20you%20are%20studying

Van Der Kaap-Deeder, J., Wouters, S., Verschueren, K., Briers, V., Deeren, B., and

Vansteenkiste, M. (2016). The Pursuit of Self-Esteem and Its Motivational Implications. National Library of Medicine, Journal of the Belgian Association for Psychological Science. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5854109/

Western Governor’s University. (2023). How Social Learning Theory Works in Education.

Western Governor’s University, Texas. https://www.wgu.edu/blog/guide-social-learning-theory-education2005.html#close7

Zhao, Y., Zheng, Z., Pan, C., and Zhou, L. (2021). Self-Esteem and Academic

Engagement Among Adolescents: A Moderated Mediation Model. Frontiers in Psychology. https://www.frontiersin.org/articles/10.3389/fpsyg.2021.690828/f