The Relationship Between High School Grades Percentile Ranking & Number Of Marriages: A Data Analysis Report
Photo courtesy of Getty Images
Florida Gulf Coast University, Department of Psychology
Introduction
The data set I chose to do my data analysis report on is the Wisconsin Longitudinal Study. I have always been attracted to the concept of longitudinal studies, specifically ones that document an individual’s course of life. The two variables this report will be focused on are number of marriages and high school grades percentile ranking. I am interested to see if the academic behaviors/habits someone had as a teenager have any relationship with their total number of marriages.
The Wisconsin Longitudinal Study is a long-term study that follows the lives of 10,317 randomly sampled people who graduated from Wisconsin high schools in 1957. The study consists of both men and women alike. All reports are from the respondents themselves and from the respondents’ parent, sibling, spouse, and sibling’s spouse. The research question I aim to answer with the data provided is if there is a statistically significant relationship between high school grades percentile ranking and number of marriages.
This study agrees with The American Psychological Association’s (APA) Ethical Principles of Psychologists and Code of Conduct because it maintains the participant’s privacy and confidentiality. As stated in 4.01 Maintaining Confidentiality, “Psychologists have a primary obligation and take reasonable precautions to protect confidential information obtained through or stored in any medium.” This study will not jeopardize the privacy or the animosity of the respondents. No information published in this report can be traced back to them.
One ethical issue that may arise while analyzing and reporting statistical data is the fact that an individual’s GPA, grades, and test scores are considered private information and are protected by the Family Educational Rights and Privacy Act (FERPA). Publishing confidential information without direct consent could be an infringement on their rights. To combat this, the Wisconsin Longitudinal Study’s staff has taken the appropriate steps to disguise its participant’s identities. No personally identifiable information will be published in this study.
To ensure my personal data analysis and reporting align with the APA’s Ethical Principles of Psychologists and Code of Conduct, I will make sure to respect the privacy of the study’s participants. I will only discuss confidential information for appropriate scientific purposes.
Hypothesis and Descriptive Data Analysis
H0: There is not a statistically significant relationship between a person’s high school grades percentile ranking and number of marriages.
H1: There is a statistically significant relationship between a person’s high school grades percentile ranking and number of marriages.
The Wisconsin Longitudinal Study has a large sample size with n= 9,105. This is less than the original sample size of 10,317 because “not ascertained” responses will not be included in the final calculations. Nonetheless, this is still a large sample size with a substantial amount of data points to work with. With a large sample size like this, researchers can be confident in generalizing and applying their findings to the general public population.
It is also important to note the scale of measurement. Both variables are discrete, with number of marriages being ordinal while high school grades percentile ranking is interval. When conducting research analysis, the type of data determines which statistical procedure should be implemented.
The statistical procedure chosen to help answer the research question is the Pearson Correlational test. Since we are looking for a relationship between high school grades percentile ranking and number of marriages, we must test the correlation. Whether it be a positive, negative, or no correlation at all, this test will allow me to examine the relationship between these two variables.
I am choosing to run my data through a p-value test with the set alpha as .05 in order to determine if my data is statistically significant and not just due to chance. If the p-value score is less than or equal .05, I can be confident that my data is accurate and not attributable to chance factors. The statistical program JASP will be used for all calculations.
High School Grades Percentile Ranking: x̅ = 47.71 σ = 30.82
Number of Marriages: x̅ = 1.012 σ = .374
The average high school grades percentile ranking is greater than the average number of marriages. The difference between the two is 46. This is mainly due to the variable’s different scales of measurement.
The data appears to be normally distributed for both the high school grades percentile histogram and the number of marriages histogram. This means that the data points are evenly dispersed, not skewed. However, it is still important to note that there is not much deviation in the high school grades percentile’s histogram. This is apparent in the lack of a defined curve/shape.
Results and Inferential Data Analysis
After running a Pearson correlation in JASP, I did not find statistically significant evidence of a relationship between high school grades percentile ranking and number of marriages. The correlation coefficient, or r value, must be at least .2 away from 0 for any correlation to be considered. The calculated r value was .075, which is very close to 0. This means there is little to no association between these two variables.
Through data analysis, I was able to determine that these results are statistically significant. To rule out any chance factors and other potential oddities, I ran the data through an additional p-value test. The p-value < .001, which is an incredibly low number. Since this is less than the set alpha .05, we can be confident about the accuracy of these results.
This proves that the results found were statistically significant and the data provides evidence for a valid effect. Researchers can safely conclude that there is no relationship between an individual’s grades in high school and number of marriages they have been in.
Conclusion
Based on the results from the data, I determined there is no statistically significant relationship between high school grades percentile ranking and number of marriages. These two variables have little to no association with each other and, consequently, are not correlated. A person’s high school grades in no way indicates the number of marriages they might have.
I chose the Pearson Correlational test as the data analysis procedure because I was looking for a correlation between high school grades percentile ranking and number of marriages. I ran a p-value test to eliminate the possibility of any chance factors, setting the alpha as .05 because I wanted a standard test. The visual aids provided are critical in order to fully understand the findings of this report.
I think it would be appropriate to conduct more research on this topic. Although I did not find a statistically significant relationship between the variables, other researchers could use different statistical procedures and uncover something else entirely. It could be as simple as changing the set alpha or even adding in another variable to create a mixed group design. Either way, there is still room for interpretation.
I believe that the results found through this data analysis report would be most useful in an academic setting. High schools across America could use the information to help students, specifically those struggling with depression. I would advise staff to spread a message of positivity and support, emphasizing that their current academic standing does not accurately represent the person they will become.
If published, these results will also help future researchers on this topic. If someone was also looking for early behavioral indicators to predict the likelihood of a failed marriage, they would know that it is not related to grades after some preliminary research. This way researchers can direct their attention to another variable and not waste valuable time and resources.
Similar studies have been conducted to find causes and early indicators of divorce. In a research article by Amit Kaplan and Anat Herbst titled Stratified patterns of divorce: Earnings, education, and gender examined the divorce rates in Israel. What they found was insightful, explaining that “divorce rates are much lower when both spouses have a university education (7%), whereas for all other couples, more than 10% of marriages ended in divorce.” (p 960). This statement implies that education does play a role in divorce rates. They continue discussing the results, saying, “The highest percentages (12.5%) are among couples in which the wife has a non-academic degree and the husband is less educated.”
Although education does play a role in divorce rates, it seems to be linked more to higher education, such as university, and socioeconomic positioning. This makes sense because education and wealth often go hand in hand.
Looking back through more historic studies, we can see the rate of divorce and number of marriages slowly increase. In Perceived Causes of Divorce: An Analysis of Interrelationships by Margaret Guminski Cleek and T. Allan Pearson, readers can clearly see this striking pattern unfold. “Of those participants filing for divorce in 1980, 90.7% had not been married previously; in 1981, 80.3% were previously unmarried.” (p 180). This sharp uptick occurred within the span of one year. This data implies that although most divorces happen in the first marriage, but that statistic is steadily changing, and it may not be the case for long.
There are countless variables that can influence and lead to divorce. Studies such as these offer insight into the most intimate relationships of the human experience. It is my belief that education is only one piece of the puzzle.
Figure 1
Histogram of High School Grades Percentile Ranking
This figure shows the frequency distribution of high school grades’ percentile ranking.
Figure 2
Histogram of Number of Marriages
This figure shows the frequency distribution of number of marriages.
Figure 3
Box and Whisker Plot of High School Grades Percentile Ranking
This figure shows a five-number summary of the data. This includes the minimum, first quartile, median, third quartile, and maximum.
Figure 4
Heat Map of Variables
This figure shows a correlational heat map between high school grades percentile ranking and number of marriages.
References
American Psychological Association. (2017). Ethical principles of psychologists and code of conduct (2002, amended effective June 1, 2010, and January 1, 2017). https://www.apa.org/ethics/code/
Cleek, M., & Pearson, T. (1985). Perceived Causes of Divorce: An Analysis of Interrelationships. Journal of Marriage and Family, 47(1), 179-183. doi:10.2307/352080
Kaplan, A., & Herbst, A. (2015). Stratified patterns of divorce: Earnings, education, and gender. Demographic Research,32, 949-982. Retrieved June 25, 2021, from http://www.jstor.org/stable/26350140