OOS\chap5

Computers Byte Back - Chapter 5, OOS Survey Methodology

Chapter 5
OOS Survey Methodology

5.1 Introduction

This chapter gives details of a survey carried out to examine the current status of OOS in New Zealand IT companies and organisations. The questions contained in the survey were based on the perceived state of OOS (as found in the literature review). After briefly examining the motivation behind the survey, this chapter describes the original objectives of the survey. Explanations of the questions devised to fulfil these objectives are then listed. Finally, the survey methodology and its inherent weaknesses are given. Note that a copy of the survey can be found in Appendix A.

5.2 Motivation

The previous chapters have contained information on both how OOS is perceived in the literature, and on what measures should be taken to prevent it. It is, however, a difficult task to ascertain whether or not the literature (often sourced from outside New Zealand) reflects the state of OOS in New Zealand. OSH has written widely on the causes of OOS and how it should be prevented, yet to the author's knowledge, there have been no in-depth studies undertaken to ascertain whether or not OSH's recommendations are actually being followed. Hence this survey set out to gain an understanding of the state of OOS in New Zealand. More specifically, it attempted to gauge the understanding and acceptance of OOS in the management of the IT industry. It sought to discover what measures organisations are taking to prevent OOS, as well as ascertaining the costs OOS inflicts on the organisation. A survey carried out in the United Kingdom in 1992 will allow comparisons to be drawn between the state of OOS in New Zealand and the UK [Khilji and Smithson, 1994]. The objectives of the survey are listed on the following page.

5.3 Objectives

The main objectives of the survey were as follows:

To determine the level of awareness that New Zealand businesses have regarding OOS.
To examine the effectiveness and levels of use of software relating to OOS.
To examine what measures the businesses have taken to prevent/alleviate OOS (with special attention to the OSH code).
To discover IT managers' attitudes to OOS and its prevention.
To examine the costs OOS inflicts on a business.
To gain an understanding of the true incidence of OOS in the New Zealand IT industry.

5.4 Structure

The survey was divided into five sections. Each of the sections covered a different aspect of the organisation and are briefly described below:

5.4.1 Business Understanding (Section 1)

This section contained questions relating to the background of the business. The questions sought information relating to the number of computer users in the organisation, and how much time they spent using computers. It also asked questions relating to the organisation's finances, specifically the net profit and their average salary. These figures were required to enable comparison between each organisation.

5.4.2 Office Environment (Section 2)

This section aimed at discovering what physical measures (including furniture and computer hardware) the business had taken to prevent OOS. It began with several questions that checked to see if the organisation followed the recommendations contained in the OSH code. For example, it questioned whether the computer users' chairs were adjustable (as recommended by OSH). After these preliminary questions, the survey examined the organisation's views on ergonomic equipment, and the costs involved with the purchasing of this equipment.

5.4.3 Education of Employees (Section 3)
This component of the survey endeavoured to discover what methods the organisation used to educate their employees about the risks of OOS. It also included questions relating to the costs the organisation incurred in providing their education programmes.

5.4.4 Management Awareness of OOS (Section 4)

This section covered aspects concerning the managerial awareness and perception of OOS. It contained questions relating to the working environment (for example, Does the business monitor employees' keystrokes [as a measure of productivity]? Are employees encouraged to notify management if they develop OOS symptoms? ); the manager's personal views on OOS; the number of cases of OOS in the organisation; and whether or not the organisation has an OOS policy.

5.4.5 Provision of Preventative Software (Section 5)

The final section of the survey covered the organisation's awareness of break reminder tools and exercise software. This software is described in the literature [Glatter 1996; Jones and Mattinson 1994], but its effectiveness has never been proven or disproven. Hence this component of the survey examined how widespread the use of such software is.

5.5 Survey Methodology

The survey was combined with two other surveys from the fourth year Information Science honours class, forming a survey of twenty pages in length. Although the resulting survey was lengthy, it was hoped that it would be looked on more favourably than three separate surveys. Last year there were complaints from several organisations that had received more than one survey from the University - it was hoped that the approach of combining the surveys would prevent the University receiving similar complaints.

5.5.1 Selection of Businesses

A database containing the addresses of approximately 1000 businesses was used as a basis for determining the organisations that were to be surveyed. The sample was made up of a

Dr. Stephen MacDonell, a senior lecturer at the University of Otago, kindly permitted us the use of the database he had used as part of his academic research. 30

cross-section of New Zealand companies and organisations with involvement in the IT industry. We examined the New Zealand Business Who's Who to ascertain whether or not all of the businesses were still in operation. After removing all of the businesses that did not appear to be operating, we had a sample of 841 businesses and organisations.

5.5.2 Pilot Survey

In late May, pilot surveys were sent out to twenty-eight Dunedin businesses. The pilot survey enabled us to check whether our questions were reasonable, unambiguous and answerable. After the pilot surveys were returned, we analysed the responses, and made appropriate (but minor) changes to our surveys (note that the cover letter to the pilot survey can be found in Appendix B(i)).

5.5.3 Survey Replies

In mid-June we sent out 841 surveys nationwide by mail. The survey was addressed to the IT manager of the organisation. With each survey we included an accompanying cover letter (see Appendix B(ii)) which detailed the motivation behind the survey and a pre-addressed free post reply envelope. Due to a low response rate, we sent out reminder letters to all of the organisations in mid-July, whether they replied or not (see Appendix B(iii)). The reason for sending out reminder letters to everyone was that, due to the anonymity of the survey, we did not know who had and had not replied. Fifteen organisations replied to the reminder letter requesting a new copy of the survey. These organisations were forwarded another copy of the survey (see Appendix B(iv) for a copy of the letter that accompanied the resent surveys).

Figure 4: A pie chart showing the make-up of the survey replies

The cut-off dates for replies was set at September 1

, and any replies received after this date were discarded. At this time we had received 122 surveys on OOS completely filled out and 12 replies where certain questions or sections had been omitted.

Figure 4: A pie chart showing the make-up of the survey replies

This left 707 surveys left unanswered. Thus, as Figure 4 illustrates, 16% of the organisations replied.

5.5.4 Limitations in the Survey Methodology

The major limitation in this survey was that the response rate was too low to allow representative conclusions to be drawn. This was most likely due to the fact that the length of the survey was prohibitive. Because of the low response, the results cannot be generalised to represent the New Zealand IT industry.

Other limitations in the survey methodology included:

As with all surveys, the respondents may not have completed the survey accurately, for instance they may have guessed (or approximated) some figures.
The managers may have felt threatened by the intrusiveness of the questions and not answered the questions truthfully.
Despite a pilot survey having been carried out (to check the clarity of the questions), some respondents may have found the questions ambiguous, thus causing the answers to be inconsistent between surveys.
The original sample of organisations and companies to be surveyed was not randomly chosen, hence even if every organisation and company had replied, it could not have been said that the sample was representative of the IT population.
As in Khilji and Smithson's 1992 survey, the results may be skewed by a non-response bias [Khilji and Smithson, 1994]. For example, it may be the case that only those organisations that believed OOS was real took the time to answer the survey.

5.5.5 Efficacy of the Survey

Despite the limitations described above, the data resulting from the survey is still useful. It provides a snap shot of the current understanding of OOS in over 100 organisations and companies in New Zealand, and although the results are not generalisable, they are still pertinent and meaningful. The actual results of the survey are discussed in the following chapter.