Practical Data Management and Analysis for Public Health

Practical Data Management and Analysis for Public Health

Name

Course Number

Date

Faculty Name

Practical Data Management and Analysis for Public Health

Assignment 4

1. Representation of gender

1. Boxplots

   Figure 1: Pre-intervention knowledge score by sex

   

   Figure 2: pre-intervention self-efficacy scale score by sex

   

   Figure 3: Pre-intervention motivation scale score by sex

   There was a difference in pre-intervention knowledge between males and females. Moreover, the data knowledge scores are skewed to the left for both genders. However, there does not seem to have a greater difference in pre-intervention self-efficacy scale score and motivation scale score between both genders.

2. Cross-tabulation

   		What is your sex?
   		Female	Male
   		Count	Count
   An indicator of high LDL PRE	Not High	15	22
   	High	56	107

   There were no participants with optimal and near-optimal LDL levels. However, there seems to be a difference in frequencies, High and Not High categories of LDL between females and males.

2. Race/ethnic groups

1. Boxplots


Figure 4: Pre-intervention knowledge  scale score by race/ethnicity

The pre-intervention knowledge scale scores were dissimilar between the racial categories.



Figure 5: Pre-intervention self-efficacy scale score by Race/ethnicity

The pre-intervention self-efficacy scale scores were not much difference between the racial categories.



Figure 6: Pre-intervention motivation score by race/ethnicity

The pre-intervention motivation scale scores were not different between the racial categories.

2. A cross-tabulation of race/ethnicity by high vs. not high LDL

		An indicator of high LDL PRE
		Not High	High
		Count	Count
Race/ethnicity	Non-Hispanic Black	12	49
	Non-Hispanic White	8	44
	Non-Hispanic Asian	4	21
	Hispanic/Latino	10	42
	Non-Hispanic Other	3	7

There are significant differences in frequencies of the racial categories between high and not high LDL categories.

3. Correlation between LDL levels and the scores

There are moderate negative correlations between LDL levels and pre-intervention knowledge, self-efficacy and motivation scale scores. Therefore, we conclude that the theory of this score contributing to dietary behaviours is true. Moreover, low scores are linked to higher levels of cholesterol.

4. Patterns of correlations between knowledge, self-efficacy and motivation scale scores

The three scale scores are positively correlated. Knowledge and self-efficacy have the lowest correlation, then self-efficacy and motivation scale score and the highest correlation are between knowledge and motivation score.

5. Randomisation

There are approximately equal experimental groups in each ethnic group except for the Hispanic and Non-Hispanic other.

Figure 7: Gender by the experimental group

There are approximately equal experimental groups in both genders. The findings are surprising how possible it could have been to allocate balance the experimental groups between the sex and ethnic categories.

6. LDL and experimental group

1. Summary statistics

		LDL in mg/dL PRE
		Count	Mean	Median	Standard Deviation
Group assignment	Control	105	180.58	179.28	21.14
	Treatment	95	178.07	177.10	20.73

According to the mean, median and standard deviation of LDL between the experimental groups, the assignment process was unbiased randomized.

2. Boxplot


Figure 8:LDL scores boxplots by  Experimental groups

3. Cross-tabulation

		LDL category PRE
		Optimal	Near-Optimal	Borderline High	High	Very High
		Count	Count	Count	Count	Count
Group assignment	Control	0	0	17	54	34
	Treatment	0	0	20	48	27

There is no much difference in frequencies in the experimental groups in the LDL categories.

7. Assignment of volunteers in the experimental groups in terms of in terms of their knowledge, self-efficacy, and motivation scale scores

Figure 9:Knowledge scale scores by Experimental groups

Figure 10: Self-efficacy scale scores by Experimental groups

Figure 11: Self-efficacy scale scores by Experimental groups

There is no much difference between the experimental groups in terms of the median (the measure of central tendency). However, there is some difference in distributions.