Data Collection

Insufficient sleep is regarded as a public health hazard. The previous researchers have associated the lack of sleep to road carnages, industrial accidents, medical as well as other occupational errors. (3) This study is therefore designed to highlight the extent to which quality of sleep affects the stress level and performance of IT employees. It’s believed that assisting employees to reduce sleep disruptions is one way of boosting their job impact and hence improve the production quality.

The outcome of research work is largely dependent on the data collected and analyzed. Prior to carrying out qualitative and quantitative data analysis, the data sources need to be identified. Afterwards the data is collected and documented. The stored data is later analyzed, and conclusions are driven based on research objectives.

The focus of the research was to evaluate the impact of sleep on the IT employees in the USA. The data was collected from the employees in the IT field, psychology experts and CEO of IT companies in the USA. To narrow down the data collection to the research level 4 IT companies were identified at random across the country whose employees and CEO provided the data. In addition, we identified 3 psychology experts who also assisted in data provision.

The table below was used as a guideline during the data collection process.

Table 1: Data collection guideline

Name of data source	Organization	Data description	Data file format	URL	Fee	Target data source
Employee’s questionnaires		Questionnaire result forms	txt		Free	Yes
Interviewing CEOs		Interview result	Txt		Free	Yes
Report		Psychology expert report	Txt		Free	Yes

The data was collected via the use of questionnaires and face to face interviews

• Storage of Data

Upon collecting data from an identified source, the results were tabled and stored in folders.

Afterwards the table below was used to document the data stored so that all the details are taken in to account during the analysis phase.

Table 2: Data Storage

Data source name	Date of collection	Location of the file saved	Filename	Format of the saved file	No of records
CEO response	3/Jan/2026	//raw data/	Questionare.txt	txt	40
Employees response	23/March/2036	//raw data/	Interviews.txt	txt	56
Expert response	6/June/2016	//raw data/	Report.txt	txt	60

1. Designing and Implementation
2. Pre-Processing of Data

Since not all the people do respond to the questionnaires and those who respond occasionally give erroneous information, it’s important to do data pre-processing. This will involve reading the raw data collected from the study, filtering the duplicated data and void data, resembling the data to derive new data and finally recording the new data in a file.

The aim of the data pre-processing is to come up with information whose quality can be guaranteed. The pre-processing will therefore focus on eliminating repetitive information, scraping of biased or void data, eliminating information that can be viewed as misleading or misrepresenting the actual situation.

After completing the data pre-processing the next task is to derive features from the results and minimize the number of random data which are to be considered. (6) The table below shows the result of the pre-processing and reduction of random data. The new file created will now be applied in conducting research analysis.

Data Analysis

Table 3: Structure of Dimension reduction

Date	Data source name	Objective of pre-processing	Method of pre-processing	Original records	Newly obtained data record	Name of the new data file
25/7/2016	CEO response	Eliminating duplication	Data cleaning	150	100	Final_CEO_response.txt
25/7/2016	Employees response	Fill the missing parts	Data filtering	96	70	Final_employee_response.txt
25/7/2016	Expert response	Eliminating repetition	Data cleaning	88	65	Final_expert_response.txt

• Experiment design

The experiment was based on a hybrid methodology which catered for both the qualitative and quantitative research approach. The data was collected and then analyzed to give answers to some of the predetermined questions. the areas of focus were the work overload, conflict of roles, work stress and fatigue. The outcome of the research was then recorded as statistics.

1. Experiment implementation records.

The data were analyzed using the Microsoft Excel. The information was then presented in the form of tables, figures and graphs and conclusions drawn.

Table 4: Frequency of sleep problems

Variable	Mean
Difficulty initiating sleep in the past month	5.34
Problems maintaining sleep in the past month	6.59
Non-restorative sleep in the past month	5.00

Table 5: Work stressors

Variable	Mean
Work overload	2.7
Conflict of roles	2.3
Repetitive work	3.29
Autonomy of job	3.23
Symptoms of depression	2.75

Table 6:Gender

Variable	%
Female	52.77
Male	47.23

Table 7:Marital status

variable	%
Married	55.45
Separated	16.79
Single	27.76

Figure 2: A pie Chart of Marital Status

Table 8:Level of education

Variable	%
High school and below	27.29
College	28.57
Undergraduate	24.43
Past undergraduate	19.71

Table 9:High stress

Variable	%
Good sleepers	27
Poor sleepers	65

1. Result Analysis and summary
2. The expected results

This study contributed to the growing literature concerned by the stress of employees and poor-quality sleep. (8) The area of sleep quality has not been effectively explored with most research either using insufficient samples to generalize or rely on convenience samples. This study however has focused on the full time IT employees hence drawing on a large sample to make specific conclusions to the IT field. The observations can be generalized that those who get enough sleep often have minimal stress. The analysis put good sleepers at 27% when it comes to probability of suffering from high stress, while the poor sleepers have a 65% probability of high stress. (2) The relation between work stress and the quality of sleep is more complicated than imagined. The findings that prove that role conflict and work overload are associated with poor sleep are backed by other studies conducted in national contexts.

Having proven that quality sleep can improve the employees’ performance, the employers need to take on measures that may assist their workforce to improve their sleep quality. (10) one way of initiating this is to solve employees stress that is originating from work. There exist several sleep programs whish the employers can encourage their workers to enroll in. this is more useful especially for the workers who find it hard to initiate or maintain quality sleep. Even though many programs are more concerned with exercise and nutrition there are others designed to assist those with sleeping issues. (4)

The programs can advise the employees on the benefits of sleep as well as lifestyles to maintain as a way of improving the sleep quality. Employers by addressing this pressing issue will not only boost the overall employees’ productivity but also show care for their welfare. This is a way of also increasing the employees’ loyalty. With this research people can now find a basis for taking up cause of sleep health in their private lives as well as trigger the topic of sleep health in the workplace. (1)

1. Outline of the Experiment and Analysis of Results
2. Data collection
3. Data sources
4. Data storage
5. Designing and Implementation
6. Pre-Processing of Data
7. Dimension Reduction

• Experiment design

1. Experiment implementation records.
2. Result Analysis and summary
3. The expected results
4. Result Summary

References

1. Horizons Workforce Consulting. The Impact of Sleep on Work Performance and Quality of Life. Horizons Workforce Consulting; February 2015. Available from: https://solutionsatwork.brighthorizons.com/~/media/BH/SAW/PDFs/Consulting/2015-HWC-Sleep-Study.ashx
2. Akerstedt T. Shift work and disturbed sleep/wakefulness. Occupational Medicine. 2003;53:89–94.
3. Akerstedt T, Knutsson A, Westerholm P, Theorell T, Alfredsson L, Kecklund G. Sleep disturbances, work stress and work hours: A cross-sectional study. Journal of Psychosomatic Research. 2002b;53:741–748.
4. Aneshensel CS, Frerichs RR, Clark VA, Yokopenic PA. Measuring depression in the community: A comparison of telephone and personal interviews. The Public Opinion Quarterly. 1982;46:110–121.
5. Barron D. The analysis of count data: Overdispersion and autocorrelation. Sociological Methodology. 1992;22:179–220.
6. Belkic KL, Landsbergis PA, Schnall PL, Baker D. Is job strain a major source of cardiovascular disease risk? Scandinavian Journal of Work, Environment and Health. 2004;30:85–128.
7. Cropanzano R, Goldman BM, Benson L., III . Organizational justice. In: Barling J, Kelloway EK, Frone MR, editors. Handbook of work stress.Sage; Thousand Oaks, CA: 2005. pp. 63–87.
8. Ducharme LJ, Martin JK. Unrewarding work, coworker support, and job satisfaction: A test of the buffering hypothesis. Work & Occupations. 2000;27:223–243.
9. Edell-Gustafsson UM, Kritz EI, Bogren IK. Self-reported sleep quality, strain and health in relation to perceived working conditions in females. Scandinavian Journal of Caring Sciences. 2002;16:179–187.
10. Gislason T, Almqvist M. Somatic diseases and sleep complaints. Acta Medica Scandinavica. 1987;221:475–481
11. Leigh JP. Employee and job attributes as predictors of absenteeism in a national sample of workers. Social Science & Medicine. 1991;33:127–137.
12. Mirowsky J, Ross CE. Control or defense? Depression and the sense of control over good and bad outcomes. Journal of Health and Social Behavior. 1990;31:71–86.