The goal of this assignment is to measure the level of self-compassion as well as the self-care of BSW and MSW students in a Social Work Program at a regional University. I will be using the following two reliable and validated instruments to measure their level of self-compassion and self-care as they immerse themselves in the helping profession. I hope to see how the SC of the students correlates to other independent variables i.e. undergrad/grad program social work, age, education level, religiosity, spirituality, gender, etc.
The chart below illustrates the data management workflow to create an analytic data set.
The original survey data has three components:
These three components have different portions of missing values. I split the original data set into three different subsets of data and impute the missing values related to the self-compassion and gratitude data based on the survey instruments. Since there are only a missing values, I replace the missing values in each survey question with the mode of the associated survey item. I then create indexes of the two instruments separately to aggregate the information in the two data sets.
I will perform both principal component analysis (PCA) and exploratory factor analysis (EFA).
This instrument contains only the data associated with the 12 items in the survey instrument. In the original data file, the 12 variables are names Q2_1, Q2_2, …, Q2_12. We impute the missing value by replacing the missing value in each of the 12 items with the mode of the corresponding survey items. Since there are only a few missing values in this instrument, this imputation will not impact the subsequent PCA and EFA.
The gratitude questionnaire contains only the variables associated with gratitude questions. The variables used in the original data file are named Q3_1 through Q3_6. I use the same mode imputation method from above to fill in the missing vales as used in the above self-compassion survey data. The gratitude questionnaire also has minimal missing values, thus the imputation will not impact the subsequent PCA and EFA.
Since the Likert scales of Q3_3 and Q3_6 were in reverse order, I transform the order back to the original order.
The demographic variables are suffering from both missing values and imbalanced categories. About 15 records in the data sets are missing the demographic information. Therefore, these records were deleted from the final data used for analysis.
A few missing values occurred in the years of education and employment that are imputed using the auxiliary information in the variables of age, the years of education, and the length of employment.
I redefine the demographic variables below.
Question 9 asked about the respondents gender. Below is the list of their options and the corresponding answer key.
## gender.cat
## Male Female
## 7 94
## Transgender Female Transgender Male
## 1 0
## Gender Varient/Non-Conforming Prefer not to Answer
## 1 0
## Agender
## 1
Question 11 asked about the respondents race/ethnicity. Below is the list of their options and the corresponding answer key.
## race.cat
## Euro-American/White African American/Black
## 95 5
## Native American/American Indian Asian
## 0 0
## Hispanic/Latinx
## 4
Question 13 asked about the respondents marital status. Below is the list of their options and the corresponding answer key.
## marital.st.cat
## Single Married/Civil Partner Divorced
## 52 29 2
## Separated Widowed Partner
## 2 0 19
Question 14 asked if the respondent regarded themselves as someone with a disability of any kind. Below is the list of their options and the corresponding answer key.
## disability.cat
## Yes No
## 23 81
Question 15 asked about the respondents religious affiliation. Below is the list of their options and the corresponding answer key.
## religion.cat
## Protestant Catholic
## 4 9
## Baptist Methodist
## 3 0
## Orthodox Christian
## 0 1
## Jewish Buddhist
## 6 10
## No Religion Non-Denominational Higher Power
## 46 5
## Witchcraft Prefer not to answer
## 17 0
Question 16 asked about the respondents sexual orientation. Below is the list of their options and the corresponding answer key.
## Sexual.orient.cat
## Gay Lesbian Bisexual
## 1 2 13
## Heterosexual Pansexual Queer
## 78 2 5
## Questioning/Unsure Same-gender loving Asexual
## 2 0 0
## Perfer Not to Discuss
## 1
Question 17 asked about the respondents political affiliation. Below is the list of their options and the corresponding answer key.
## poli.affil.cat
## Strong Republican Moderate Republican Liberal Republican
## 3 2 4
## Independent Conservative Democrat Moderate Democrat
## 31 2 29
## Strong Democrat
## 30
Question 18 asked about the respondents educational level at their current institution. Below is the list of their options and the corresponding answer key.
## SW.Program.cat
## Bachelor's Master's
## 41 62
Question 19 asked if the respondent lived in an urban, suburban, or rural area. Below is the list of their options and the corresponding answer key.
## Urbanity.cat
## Urban Rural Suburban
## 33 27 44
Now that we have deleted and replaced the missing data values in the methods listed an shown above, the new data set is primed and ready for further analysis.
Now, I will complete Principal Component Analysis (PCA) for the compassion and gratitude questions.
Below is a breakdown of the first 12 PCA’s for each variable. I will be utilizing PC1 for further analysis.
| PC1 | PC2 | PC3 | PC4 | PC5 | PC6 | PC7 | PC8 | PC9 | PC10 | PC11 | PC12 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Q2_1 | 0.29 | -0.44 | 0.22 | -0.09 | -0.23 | 0.14 | -0.03 | 0.05 | 0.56 | 0.18 | -0.05 | -0.49 |
| Q2_2 | -0.28 | -0.34 | -0.30 | -0.07 | 0.45 | 0.33 | -0.09 | -0.08 | 0.43 | -0.18 | 0.28 | 0.29 |
| Q2_3 | -0.27 | -0.24 | 0.11 | 0.53 | -0.20 | 0.43 | -0.07 | -0.44 | -0.26 | -0.14 | -0.21 | -0.13 |
| Q2_4 | 0.28 | -0.36 | -0.25 | -0.35 | 0.44 | 0.02 | -0.03 | -0.17 | -0.51 | 0.05 | -0.17 | -0.32 |
| Q2_5 | -0.25 | -0.28 | -0.32 | 0.11 | -0.15 | -0.50 | 0.63 | -0.15 | 0.06 | -0.11 | 0.07 | -0.15 |
| Q2_6 | -0.32 | -0.17 | -0.21 | 0.20 | -0.02 | 0.18 | 0.03 | 0.58 | -0.21 | 0.60 | 0.07 | -0.08 |
| Q2_7 | -0.21 | -0.29 | 0.61 | 0.17 | 0.42 | -0.32 | -0.04 | 0.33 | -0.05 | -0.25 | -0.13 | -0.02 |
| Q2_8 | 0.32 | -0.32 | 0.28 | 0.12 | 0.01 | -0.10 | 0.15 | -0.29 | -0.11 | 0.46 | 0.26 | 0.54 |
| Q2_9 | 0.33 | -0.20 | -0.07 | -0.03 | -0.19 | 0.34 | 0.40 | 0.42 | -0.09 | -0.37 | -0.30 | 0.35 |
| Q2_10 | -0.24 | -0.39 | -0.02 | -0.38 | -0.52 | -0.19 | -0.46 | 0.07 | -0.20 | -0.17 | 0.12 | 0.18 |
| Q2_11 | 0.36 | -0.02 | -0.09 | 0.36 | -0.03 | 0.00 | -0.11 | 0.19 | -0.19 | -0.33 | 0.70 | -0.23 |
| Q2_12 | 0.27 | -0.11 | -0.42 | 0.45 | 0.04 | -0.36 | -0.43 | 0.06 | 0.18 | 0.03 | -0.40 | 0.16 |
As shown below, 41.26% of the variation can be explained by the first PCA and 52.31% of the variation can be explained by the first two PCA’s.
## Importance of components:
## PC1 PC2 PC3 PC4 PC5 PC6 PC7
## Standard deviation 2.2252 1.1514 0.98661 0.90419 0.83954 0.82720 0.79413
## Proportion of Variance 0.4126 0.1105 0.08112 0.06813 0.05874 0.05702 0.05255
## Cumulative Proportion 0.4126 0.5231 0.60420 0.67233 0.73107 0.78809 0.84064
## PC8 PC9 PC10 PC11 PC12
## Standard deviation 0.73355 0.64351 0.62859 0.55488 0.50703
## Proportion of Variance 0.04484 0.03451 0.03293 0.02566 0.02142
## Cumulative Proportion 0.88548 0.91999 0.95292 0.97858 1.00000
Below is a breakdown of the first 12 PCA’s for each variable. I will be utilizing PC1 for further analysis.
| PC1 | PC2 | PC3 | PC4 | PC5 | PC6 | |
|---|---|---|---|---|---|---|
| Q3_1 | 0.44 | -0.21 | 0.20 | -0.60 | -0.18 | 0.58 |
| Q3_2 | 0.46 | -0.12 | -0.17 | -0.42 | 0.34 | -0.67 |
| Q3_3 | -0.38 | -0.56 | 0.38 | -0.07 | 0.62 | 0.09 |
| Q3_4 | 0.41 | -0.24 | -0.55 | 0.45 | 0.37 | 0.37 |
| Q3_5 | 0.39 | -0.47 | 0.45 | 0.47 | -0.37 | -0.25 |
| Q3_6 | -0.36 | -0.60 | -0.53 | -0.20 | -0.43 | -0.10 |
As shown below, 52.72% of the variation can be explained by the first PCA and 66.82% of the variation can be explained by the first two PCA’s.
## Importance of components:
## PC1 PC2 PC3 PC4 PC5 PC6
## Standard deviation 1.7786 0.9132 0.8335 0.73805 0.66660 0.56481
## Proportion of Variance 0.5272 0.1390 0.1158 0.09079 0.07406 0.05317
## Cumulative Proportion 0.5272 0.6662 0.7820 0.87277 0.94683 1.00000