Essay: Analysis Of Transport Behaviors Among Older Teenagers And Regression Analysis Of Self-reported Sedentary Hours In NSW.

Read the paper Ward, A. L., McGee, R., Freeman, C., Gendall, P. J., & Cameron, C. (2018). Transport behaviours among older teenagers from semi?rural New Zealand. Australian and New Zealand journal of public health, 42(4), 340-346.

Critically appraise of the statistical material in this paper against items 10, 12-17 of the STROBE checklist. Present your review as a 400-500 word (approx.) report.

Note:

• Only review the provided paper Ward et al, 2018. Do not read any other papers.
• Restrict your review to how well Ward et al have documented their statistical methods – that is, items 10, 12-17 of STROBE only. You may not have to address every item; just describe the major strengths and weaknesses of the authors’ descriptions of their statistical methods and results.
• For each important STROBE item:
  • state whether you believe the STROBE item is met or not,
  • support your judgment with proof or examples from the paper, and
  • describe why this inclusion or exclusion is important / how it will impact on the reader’s understanding and decision making.
• The 400-500 words is a guideline not a rule. There are no penalties for exceeding this guideline.
• There are no marks for adding a reference list. Referencing is optional.

Question 2 Note: Students will get different answers as the data sets differ.

Using R Commander and your assigned data set, write a statistical report addressing the following research question:

In the population of 17-year-old residents of NSW, does the number of activities attended in the past month (activities) predict self-reported sedentary hours per week (sed) after correcting for gender (sex).

Note: To answer this question, you need to use R Commander and the data set assigned to you for assignments. This data set contains the (fictitious) data from a random sample of 17-year-old residents of NSW and is the same data set as that you have previously used in Assignments 1 and 2. See ‘Description of your data set.docx’ for the descriptions of the variables.

Note: This assignment is assessing your skills, not the skills of the computer. You will need to include graphs from R Commander into your assignment but all other R Commander output will attract 0 marks and is discouraged. It is your task to identify the relevant results in the R Commander output and write these up in your assignment.

Also note:

• You should only use the variables ‘activities’, ‘sed’ and ‘sex’.
• Correcting for ‘sex’ is just including ‘sex’ in the regression model.  When sex is in the model all other variables are corrected for it.
• Documenting your analysis plan is recommended but not required. (If your analysis report is complete then your plan must have been complete also.)
• Do report the results of your descriptive analyses
  • Well labelled graphs can be copied from R Commander
  • Summary statistics and tables should be manually typed
  • Summarise the main findings of your descriptive analyses in words and describe how these findings inform your expectations and interpretation of the more complex models.
• Do report the results of your statistical inference and/or regression models
  • Any fitted regression models should be manually typed and described in the text.
  • Any hypothesis tests should contain all relevant information (use the 5 step method to be sure)
  • Any other results such as confidence intervals should be manually typed and described in the text.
  • Summarise the main findings of your regression model and statistical inference in one or two paragraphs.
• Do remember to answer the research question
  • Write a final paragraph which summarises the key findings of your analysis and your answer to the research question.
• Do check the Learning Guide for the marking criteria
• Do write your answers yourself and keep them private.

This paper seeks to critically analyze a research study on transport activities connected with youths in New Zealand (Ward, McGee, Freeman, Gendall, & Cameron, 2018).

Review against Items 10, 12-17 of the STROBE Checklist

In order to achieve the desired sample size, secondary schools were targeted in order to access respondents with the age of interest in this case. 775 respondents participated in the study.

Data analysis involved descriptive analysis and inferential analysis. On descriptive analysis, means, minimum, maximum and standard deviations were calculated and reported. On inferential statistics, independent (unpaired) sample t-tests and chi- square tests were conducted. The missing values in the captured data were included from the analysis. 82 percent of the study participants provided complete responses to the survey questions, while the remaining 16 percent had incomplete responses.

A pilot study was conducted on the same number of schools as those considered in the main study. This study was not conducted in stages. The pilot study was key in improving the data collection instruments. The response rate for the actual study was 71.5 percent (the 775 participants). Teenagers who participated in the class survey had a response rate of 77.2%, while those who took the survey at home had a response rate of 65.6%.

This response rate is high enough, thus the results obtained were reliable. From the report, the number of females with missing data was twice the number of males. This was because the male respondents chose to participate in the survey in class, rather than answer the questions at home. Survey in class had a higher response rate, since it was more convenient for the students. Answering the survey questions a home was ineffective due to fatigue.

From the study, 49 percent were male, while 51% were female. 7.9% were 15 years old, 49.2% were 16 years, 40.7% were 17 years, and 2.2% were 18 years, while 0.1% of them were 19 years old. In addition, 71.2% of the respondents were from urban areas and 28.8% from rural areas. Moreover, 85.1% of the participants were European nationals, while the remaining 14.9% came from other nations. In addition, 59.7% of the teenagers earned less than 50 dollars per month, 11.8% had an income of between 51 and 99 dollars, while .5% of them earned over 100 dollars per month.

The chi- square tests conducted showed significant associations between gender and some of the modes of transport. From the results, it was evident that more male than female students preferred using bicycles, using skateboards or riding motorcycles. Additionally, more female than male students preferred taking public or school buses or being passengers in cars. Further analysis revealed that more male students had driving licenses as compared to their female colleagues. In addition, t-test results revealed that more male students participated in sporting activities, while the females were more active in social and cultural events.

Descriptive Statistics

Question 2: Regression Analysis using R

Introduction

This section involves linear regression analysis using the R software. The research question in this case will be: Does the number of activities a student has attended in the past month predict self-reported sedentary hours per week after correcting for gender? The independent variables in this case will be will be activities attended and gender. The dependent variable will be sedentary hours. The research hypothesis is as give below:

H_0: Number of activities attended in a month does not predict the number of self-reported sedentary hours.

H_1: Number of activities attended in a month predicts the number of self-reported sedentary hours.

The obtained data has a total of 271 respondents. The effect of activities undertaken on sedentary hours will be checked after controlling for gender.

The codes used in R software are given on the appendix.

Results

The table above shows descriptive statistics for the two main variables of the study (activities and sedentary hours).The figures above show the histograms for the three variables used in the linear regression model. That is, activities, sedentary hours and gender.

The figure above shows a scatter plot between activities and sedentary hours.

The table above shows the model summary results of the regression model. 

ANOVA

The table above shows the analysis of variance results. These results are important in showing whether the selected model is significant.

Model Coefficients

The table above shows the model coefficients results, which help in showing whether activities a student involves in, and gender have an impact on their sedentary hours. The results will also give the regression equation of the model.

Interpretation of Results

The first table of results shows the descriptive statistics of the two main variables used in the regression analysis. Descriptive statistics give basic summaries of variables (Ho & Yu, 2015). From the results, the minimum and maximum number of activities was 2 and 13 respectively. In addition, the activities had a mean of 7.07 and a standard deviation of 2.291. The minimum and maximum number of sedentary hours was 4.5 and 20.4 hours respectively. The sedentary hours had a mean of 10.637 and a standard deviation of 3.0521.

The histograms on activities, sedentary hours and gender are important in showing the distribution of the variables (Silverman, 2018). From the results, all the three variables have an approximately normal distribution, as shown by the bell shaped normal curve.

Model Summary

The scatter plot between activities and sedentary hours shows a linear relationship between the two variables (Sloan & Angel, 2015). From the results, activities and sedentary hours have a negative relationship. That is, an increase in the number of activities that a student engages in leads to a decrease in the number of self- reported sedentary hours.

The model summary results show the correlation between independent and dependent variables, as well as the amount of variation explained by the independent variables. From the results, activities and gender have a combined correlation of 71.6% with sedentary hours. Additionally, activities and gender explain 71.6% of the total variation in the number of sedentary hours.

The analysis of variance was conducted with activities and gender as the independent variables, while sedentary hours were the dependent variable. This was in order to check for the significance of the model (Wiley & Pace, 2015; Saisana, 2014). From the results, the regression model was significant in showing the relationship between activities that a student engages in and the number of sedentary hours while controlling for gender, F = 141.167, p < 0.001.

The last table shows the model coefficient results. From the results, activities that a student engages in significantly predict the number of sedentary hours, t = -16.765, p < 0.001. In addition, gender has a significant impact on the number of sedentary hours, t = -10.526, p < 0.001. From these findings, there is enough evidence to reject the null hypothesis in favor of the alternative one which states: Number of activities attended in a month predicts the number of self-reported sedentary hours.

The regression equation model is as given below:

Sedentary hours = 21.791 – 1.294 (activities) – 3.727 (sex) 

Discussions

Data analysis involved descriptive and inferential analysis, whose results were reported in tables and figures. From the results, it was evident that the number of activities that a student engaged in the prior month predicted the number of self- reported sedentary hours after controlling for gender.

References

Ho, A. D., & Yu, C. C. (2015). Descriptive statistics for modern test score distributions: Skewness, kurtosis, discreteness, and ceiling effects. Educational and Psychological Measurement, 75(3), 365-388.

Ward, A. L., McGee, R., Freeman, C., Gendall, P. J., & Cameron, C. (2018). Transport behaviours among older teenagers from semi?rural New Zealand. Australian and New Zealand journal of public health, 42(4), 340-346.

Wiley, J. F., & Pace, L. A. (2015). Analysis of variance. In Beginning R. Apress, Berkeley, CA, 111–120.

Saisana, M. (2014). Analysis of Variance. Encyclopedia of Quality of Life and Well-Being Research. Canada: Springer. 162-165.

Silverman, B. W. (2018). Density estimation for statistics and data analysis. London: Routledge. 2(1), 175-193.

Sloan, L., & Angell, R. (2015). Two-way scatter plot and the UK Living Cost and Food Survey (2010): Household income and expenditure. SAGE Publications Ltd, 59-68.