Have you ever thought of conducting a survey of your membership to better understand the causes of identified problems and the effects on union members and those who use their services?
This information uses a workload survey as an example, and has been adapted from material produced by the Canadian Union of Public Employees (CUPE). Many OHS reps have found workload surveys to be a useful tool for understanding the causes of workload problems and how to address them.
A survey can help inform the union of how members are experiencing and coping with heavy workloads. It can provide input from a large number of union members, including from those who don't normally speak up. It can be an important way to reach out and get input from EEO groups in the union, such as migrant workers, Aboriginal workers, workers with disabilities, gay members, women and young workers.
The results can provide valuable information to help you and your union develop more effective strategies to tackle workload problems. A survey can gather ammunition for a campaign that calls attention to the need for more staff, demonstrating how the quality of work, and/or services to the public decline even as members work harder and longer trying to fill the gap.
The following pages provide a guide to help you figure out how to do a workload survey, including what to think about when developing questions, how to analyse the information and how you can use the results.
When preparing a survey, it is important to be very clear about its objectives and emphasis. Ask yourself the following questions to help define the scope of the survey:
- What is the purpose of the survey?
- Who will be asked to respond to the survey?
- What information do we want to have when the survey is complete? (Be very specific in answering this.)
- How will we use the information we gather?
- What are our plan and timelines for getting surveys completed and results analysed?
- What resources do we need to do this?
Asking questions and getting answers should be a very straightforward process...and it can be. With proper planning, many problems can be avoided at the start. The following suggestions should be helpful in making sure that everything goes smoothly.
1. Who has the information you are seeking?
Be sure to direct the survey to those who have the information you need. If you send a survey out to someone who does not have the information, you won't get the results you're looking for. For example, if you're looking for detailed information on how individual members experience workload problems, the survey must be completed by individual members. But, if you need information about the number of positions lost in a workplace over time, an individual may not be the most reliable source of information for this. It may be more appropriate to survey the union organisers in this case since they are most likely to have this information.
2. Developing a Questionnaire.
There are basically two kinds of questions used in a questionnaire:
- "open ended" questions which allow respondents to fill in their own answers, e.g. "What do you think are the causes of increased workload?" and
- "closed" questions where the answers are listed and respondents tick off the ones they feel match their experience, eg "What are the sources of workload pressures:
- poor management,
- funding cuts,
- contracting out,
- fewer staff, etc."
In designing a questionnaire it is important to balance the goal of creating a clear, short, "to the point" questionnaire with the need to make sure the survey collects all the necessary information. Please consider the following:
- Is this an opportunity to solicit information from groups we don't often hear from, like EEO groups? Does the issue you are investigating affect them in a different way, which might require different questions or solutions?
- Are there issues like language or literacy that need to be taken into account?
- Are the objectives of the survey clearly presented to the respondent? (The more respondents understand how the information will be used the more likely they are to respond.)
- What demographic information do we need? Gender? Age? Job classification? Employment status? Ethnicity? Why do we need this information?
- Will the responses to the questions provide background information to help interpret the results?
- Are the questions and the options for answers clear?
- Can the questionnaire be answered in 10 -15 minutes? (if not, then it's too long)
- Will the results be easy to tabulate and provide the information you're looking for?
On the last point, closed questions (tick boxes) are the easiest to tabulate but they do not provide as much opportunity for the respondents to explain their answers. Most large surveys are based largely on closed questions, but almost all ensure there is an opportunity to add written comments.When relying on closed questions, it is very important to make sure that the questions are well written and provide an opportunity for as full a range of responses as possible.
It is crucial to check the clarity of the questions by pre-testing the survey with a small group that resembles the larger group in occupation, gender, etc. The test group should complete a draft of the questionnaire and comment on any problems they found (confusing questions, missing possible responses, etc.) Also, it is often helpful to review the questionnaire with someone familiar with survey design in order to identify problems in advance.
Here are some sample questions you could include in a workload survey:
- In the past year has there been an increase in your workload?
- Don't know
- In the past year, did you ever go to work early or stay late outside of your regular or normal working hours in order to keep up with your workload?
- Don't know
- Does your workload negatively affect your family, religious or cultural responsibilities? If so, how?
- Typically, how much unpaid work do you do? (This is work for which you receive no compensation whatsoever, not even regular pay or time off in lieu of pay.)
- Less than 10 minutes a day
- 10-30 minutes a day
- more than 30 minutes a day
- more than 60 minutes a day
- More than 90 minutes a day
- Please indicate if any of the following have contributed to your workload:
- Staff reductions
- Additional job duties
- Training other employees
- Cutbacks to funding
- Lack of resources
- New technology (i.e. computers)
- Increasing paperwork
- Intensity (speed-up) of work
- Other, please specify
- Do you believe that your workload is affecting your health?
- In the past year, have you:
- Made a claim for Workers' Compensation
- Taken sick leave due to work related problems
- Does being a member of an EEO group contribute to workload problems? If so, how?
Even the best questionnaire is of no use if there are problems getting it out to the respondents and getting it back. The following suggestions may help with this crucial part of the survey process:
- Make certain the survey includes a clear explanation of its purpose and how it will be used.
- Identify a contact person in case the respondent has any questions or concerns.
- The questionnaire should also include clear directions about how it is to be returned, to whom and by what date.
- Allow lots of time for the survey to get out to respondents and to be filled out and returned. This is particularly important in cases where the respondents include part-time or shift workers, where respondents work in a variety of locations, and when the survey asks for information that might not be right at hand.
- Create a plan and a network for collecting the surveys (e.g. delegates' or shop stewards' meeting, convenient drop off points, or meetings where the survey can be filled out and collected at the same time). Make sure the plan takes into account access issues like transportation, language, security, etc.
- Follow up where the surveys have not been returned (eg with a telephone call or personal contact) - emphasizing that the better the return rate, the more representative the results. (A 25 per cent response rate is fine, as long as it is representative of the entire group.) Use this as an opportunity to invite input from under-represented groups such as EEO groups, especially if their views and experiences are not often heard.
The method of tabulating the results depends on the type of questions asked, the number of responses and the survey's purpose. For a small survey it's possible to tabulate the results manually, or with a simple data base program, like Microsoft Access. For a large, more detailed survey with many respondents it's best to use a computer program designed for analysing survey results like SPSS (Statistical Package for the Social Sciences). Your union's Research staff should be able to provide help and advice on this. Please keep the following points in mind:
- Clarify the financial and human resources you will need to analyse the survey before you begin. The technical and financial requirements of compiling the results (coding questionnaires, data analysis and interpretation, including costs and time required) must be considered when the survey is designed. It is disappointing to complete a survey only to find it cannot be analysed as expected.
- Participation in the survey is the first step in a communication process with the members. The results of the survey should be communicated back to the members for review and further discussion. Care must be taken to ensure that all communication or reports protect the confidentiality of individual respondents. Inform members of the results even if you're not making all the results public, for example, because of ongoing negotiations with management.
- Survey results may be presented in a written or verbal report or a membership newsletter. The presentation depends on the circumstances of the survey and the means of communication available. For example, survey results may be summarized in a presentation at a conference or in a newsletter. Think about literacy issues when deciding how to communicate results.
How you put your survey results to use depends on your original purpose for conducting the survey. Was it to understand better the nature, causes and consequences of increasing workloads to inform bargaining? Results can help to identify what contract changes are needed to address workload problems. In this case, results may be used internally by the union to develop strategy and to explore the different impact on EEO groups.
Survey results can also be used publicly. They can form the basis of a public report or a press release, for example, calling attention to workload problems for your members and the negative effect on the quality of services.
Conducting a survey can be an important element in a public campaign to mount public pressure for improvements. In this case a strategy for communicating the results to the public will be important to make the most of the survey. Your union should be able to provide assistance here.
- Guide to writing survey questions - From MAD (Management Analysis and Development)
- 10 Common mistakes made when writing surveys - From Surveygizmo
Last amended January 2015
Acknowledgement: Based on a CUPE (Canadian Union of Public Employees) document.