Does going to university in a different country affect your mental health? A Japanese international university surveyed its students in 2018 and published a study the following year that was approved by several ethical and regulatory boards.
The study found that international students have a higher risk of mental health difficulties than the general population, and that social connectedness (belonging to a social group) and acculturative stress (stress associated with joining a new culture) are predictive of depression.
Explore the students data using PostgreSQL to find out if you would come to a similar conclusion for international students and see if the length of stay is a contributing factor.
Here is a data description of the columns you may find helpful.
| Field Name | Description |
|---|---|
inter_dom | Types of students (international or domestic) |
japanese_cate | Japanese language proficiency |
english_cate | English language proficiency |
academic | Current academic level (undergraduate or graduate) |
age | Current age of student |
stay | Current length of stay in years |
todep | Total score of depression (PHQ-9 test) |
tosc | Total score of social connectedness (SCS test) |
toas | Total score of acculturative stress (ASISS test) |
-- Run this code to save the CSV file as students
SELECT *
FROM 'students.csv';--Explore student data - Total Records and Count Inter-student records
SELECT COUNT(*) as total_records
From students
SELECT inter_dom, COUNT(*) as total_records, COUNT(inter_dom) as count_inter_dom
FROM students
GROUP BY inter_dom;--Summary of Tests Depression, Social Connectedness and Stress. Groupby Student TYPE
SELECT inter_dom, min(todep) as min_depression, max(todep) as max_depression, ROUND(AVG(todep), 2) as avg_depression_score,
min(tosc) as min_social, max(tosc) as max_social, ROUND(AVG(tosc), 2) AS avg_social_score,
min(toas) as min_stress, max(tosc) as max_stress, ROUND(AVG(toas), 2) as avg_stress
From students
WHERE inter_dom is not null
GROUP BY inter_dom;
--Impact of Length of stay on average scores. Group by Studnet Type to see differences.
SELECT inter_dom, stay, ROUND(AVG(todep), 2) as avg_depression_score,
ROUND(AVG(tosc), 2) AS avg_social_score,
ROUND(AVG(toas), 2) as avg_stress
From students
GROUP BY inter_dom, stay
ORDER BY stay DESC, inter_dom
--Examine Year 5 Data since AVG depression score is 0
SELECT inter_dom, stay, todep, tosc, toas
From students
WHERE stay = 5
--Since there is only 1 student for year 5 and 3 total for year 5 - count number of students per year to see if there is sufficent data
SELECT inter_dom, stay, COUNT(stay) AS num_of_students
FROM students
Group by stay, inter_dom
ORDER BY stay DESC
---International Only
SELECT stay, ROUND(AVG(todep), 2) as average_phq,
ROUND(AVG(tosc), 2) AS average_scs,
ROUND(AVG(toas), 2) as average_as
From students
WHERE inter_dom = 'Inter'
GROUP BY stay, inter_dom
ORDER BY stay DESC
--OTHER FACTORS GRADUATE VS UNDERGRADUATE
SELECT inter_dom, academic, stay, ROUND(AVG(todep), 2) as avg_depression_score,
ROUND(AVG(tosc), 2) AS avg_social_score,
ROUND(AVG(toas), 2) as avg_stress
From students
WHERE inter_dom = 'Inter'
GROUP BY inter_dom, academic, stay
ORDER BY stay DESC, academic
# DISCUSSION
After exploring data to examine if the Total score of social connectedness (SCS test) and the Total score of acculturative stress (ASISS test) are predictors of international students' Total score of depression (PHQ-9 test) greater than the general population, my review of the data does not clearly support that those factors (social connectedness and acculturative stress) as predictive of the total score of depression. Nor does the data show that international students have higher scores of depression than domestic students.
The second question was to examine if length of stay impacts all three scores.
Exploring Data and Data Credibility
Data is from a peer reviewed and published study, so I can assume data is fairly credible.
1.1 Sample Size
There were a total of 286 records which is a small sample size, with a larger sample size there may be clearer correlation or predictive measures. Out of the total 286, 201 are international, 67 are domestic and 18 are NULL. Thus international student make up 75% of the sample population and domestic or the control group equal 25%.
1.2 Irregular Data Exploration
Significantly, there was a average depression score of 0 for international students whom stayed for 5 years. Exploring futher, I discovered there was only 1 international student participant that reported for staying 5 years and 2 that were domestic. These low numbers led me to check how many students per length of stay to see if low numbers were reducing accuracy of data. Years 1-4 had most of the participants for the study both international and domestic which make sense as college or grad school is typically 4 years to completion. Year 5 and after had only 1-3 participants total, and years 6-10 were only international students. Thus, data is less reliable year 5+ due to low number of participants.
Data Analsysis: Score Summary
Data was then summarized by min, max and average for the depression, social connectedness and stress scores. Data was grouped by student type: international and domestic. There were little significant differences in the average scores across the two groups except for the stress score where there was a difference of +12.72 between domestic and internationla students. However, the stress score while higher for international students did not result in an increase in the average depression score compared to domestic students. Actually, domestic students had a higher average depression score of 8.61 over 8.04 for international students.
2.1 Digging Deeper: Length of stay as a contributing factor
When analyzing the length of stay's impact on international students' depression, the results are inconclusive that there is a correlation. When looking at just the first 4 years for which we have the most participants the is an increase in depression for the first two years and then it begins to decreases in year three and four. I also looked at the scores of international students separating from those who were undergraduate vs graduate to see if the depression scores are much higher among graduate vs undergraduate students. Undergraduate students had higher depression scores than graduate students and while there was a decrees in year 3 it also had a increase in year 4.
Thus, there needs to be more analysis and further study increasing the number of participatants to see if there are correlations between length of stay and depression.