Answer

Frequency Distribution

The article ‘Frequency Distribution: Long working hours, sleep-related problems, and near-misses/injuries in industrial settings using a nationally representative sample of workers in Japan’ is a qualitative study (Yamauchi, et al., 2019).

Cross-tabulations were used to check frequency distribution of the variables across industries because the variables were categorical. Further, Chi-square and residual analysis were used to confirm the significance.

The study had a total of 18,682 participants (7,098 (38%) female and 11,584 (62%) male). The bivariate distribution showed that 78. In terms of age, 23.8% were aged between 20 to 34 years, those aged between 35 to 49 years accounted for 43.7%, while the category aged from 50 to 64 years had a 32.5% share. In terms of education years, <16 years were 43.9% while 56.1% was in the 16+ years. 87.3% of the participants worked between 35 to 60 years hours a week, while 12.7% worked 61+ hours a week. 19.6% had night shift work while 80.4% did not. On the job-related stress variable, 76.9% of the participants had experience low stress level with the rest having high stress levels. 54% of the respondents had no sleep problems and 38.9% had experienced depression. The distribution across industries was as follows: construction had 1,800 participants (9.6%), manufacturing – 3,888 (20.8%), transport/postal service – 1,289 (6.9%), wholesale/retail trade – 2,616 (14.0%), medical/health/welfare were 2,765 (14.8%) and 33.8% of the respondents were in other unlisted industries.

The study concluded that, in terms of frequency distribution, there was significance in difference in the frequency levels distribution for each factor by industry.

Pie charts can also be used to display the distribution of this data, showing level share per variable. However, pie charts are more effective for displaying a small number of frequencies while categorical variables with many levels will not effectively display on pie charts. The other options are bar charts which also show aesthetics of variable levels or categories by composition. The con of pie chart is that it does not display bivariable relationship while bar plots can be customized to display bivariate visualizations.