Survey Services uses a generic process flow for its projects and this is customized to fit the nature of the survey. We can provide services in all or any combination of these steps. The generic steps are outlined below. In addition, special issues that need to be considered for different types of surveys are outlined next.
1. Survey project planning
We work with you to develop a comprehensive and customized project plan with timelines, responsibilities and costs. The plan will cover all the steps, where applicable, that are listed below. We will also clarify objectives and uses of the survey and how it links with other initiatives you may have in place.
2. Questionnaire design and development
Using our question bank, your existing questionnaire, or starting with a blank sheet we will help you develop a valid, consistent and reliable questionnaire in line with your project objectives. This will incorporate existing research material and plans and may involve qualitative interviews and focus groups. We will assist in ensuring appropriate language, phrasing and clarity of expression.
3. Sample/population decision
If a sample is to be used, at this point the sample parameters and design are decided. If a population survey is required, we must consider how everyone will be identified.
4. Delivery methods
This is deciding how will the questionnaire by deployed - web (intranet/internet/email hyperlink) or paper (mailed, handed out, facilitated group, etc) - and the resulting activities required as a result of this decision, eg, installing software, organizing print production, data entry, delivery times.
5. Communication program
Communication programs are not designed just to inform respondents that a survey is about to happen. They should also address how to get a high response rate, how to recognize and even reward participation, how to share and debrief results and how to act on the messages that are contained in survey results. We can design an entire communication campaign to support your survey or assist you to engage your employees in the process.
6. Pilot testing and refinement
An important step that helps achieve an outstanding result. Pilot tests not only give you valuable feedback about question wording, questionnaire layout, technology effectiveness but you can use the results to identify redundant questions.
7. Questionnaire loading delivery
This involves loading respondents, preparing delivery messages, physically sending questionnaires, or organizing collection groups.
8. Collecting responses and monitoring progress
We have on-line tools that can give you instant access to amount and type of response, for web surveys. In some cases, this step may involve providing helpdesk or helpline support.
9. Data entry, if required
This is an unnecessary step if you are using respondent data entry or data entry at point of interview, eg, CATI. However, there are always circumstances when data entry is required. We can provide data entry staff for your surveys using your own software, spreadsheets, statistical packages or our own specialized survey software systems. It is your decision whether data validation is required and the amount of free-text coding or content analysis needed.
10. Analyzing results and producing reports
The amount and type of statistical or qualitative analysis required will depend on the survey type and purpose. Also, report detail and sophistication is a consideration in this step. Survey Services can merge data sets and perform longitudinal or time series analysis which may be useful for benchmark or trend analysis.
11. Providing feedback
The amount of emphasis in this step will depend on the nature of the survey. However, it is a critical step and requires careful decision-making.
12. Deciding priorities and implementing ideas
Surveys provide information for decision-making--the don't make decisions in their own right. So this is the hard work part and the reason for initiating most research.
Special considerations
Employee surveys
- It is important that current and potential employee issues are addressed in the survey so that data can be used for tracking progress, ie, longitudinal analysis. This means taking a broader view and not just working with "hot potatoes." Working off existing initiatives, conducting focus groups and working within the framework of your organization's strategic plan all help shape the survey.
- Literacy - both language and computer - are important considerations for data collection. Alternative methods/activities may need to be used to deal with this issue.
- The level of trust within the organization is a critical issue. How willing will employees be to respond and to respond truthfully. This will affect your communication strategy.
Multi-rater surveys
- Decide the purpose of the feedback early in the planning process . This will significantly shape all other decisions in the survey process.
- Engagement (briefing respondents; having others prepared to give feedback and having employees prepared to receive feedback) is a critical success factor in this type of survey.
- Logistics - getting the right details, respondent relationships, distribution and nearly 100% response rate - are all important.
Customer satisfaction
- Distinguish between customers and stakeholders - they are usually not the same and will have quite different issues.
- Decide on the best way to contact your customers - self-completed paper or web questionnaires, face-to-face or telephone interviews, focus groups, etc.
- Focus on where you fit in terms of your customers' delivery frameworks, not where customers fit within your organization's process structure.
Outsourcing evaluation
- How will results be benchmarked and interpreted in light of performance contracts?
- How to reflect performance indicators in a valid, reliable and simple questionnaire.
- Deciding whether to weight responses of critical users.
Remuneration
- What information will you collect out of basic pay, bonuses and benefits, annuities (superannuation/pensions), vehicles, equity plan, performance pay and government taxes?
- Continuous, biannual or annual polling.
- Validation during data collection, eg, 8% annuity to be entered as 8 and not 0.08 or $1,600 (of $20,000), etc
- What calculations do you need to perform before and after data collection?
- Data collection method - interviews or questionnaires (paper, web or spreadsheet)?
- Will company size, location, industry, and turnover influence your results?
Benchmarking
- What are you really benchmarking - financial performance or service delivery?
- For sampling across organizations, ensure you use language and terms familiar to all.
- How much derived data can you ask for, ie, how much calculation is it reasonable to expect respondents to do before answering your questions, eg, cost/area, NPV of investment after 18 months, average rate of return, etc?
Inventory and asset tracking
- Are the items coded consistently, eg, furniture is 6 alphanumerics, car parts are 16 numerics, and consumables are 8 alphas?
- Distinguish between short-term inventory and raw materials.
- Are the assets truly tangible?
- How are assets with no value/fully depreciated handled, ie, where is the cut-off decision made?
Process control
- Is data obtained from people or machines?
- What is being measured - qualitative or quantitative data?
- How will out-of-process range data be handled?
- What type of trend analysis will be used to identify process runs?
- Consider the issues of sampling frequency, accuracy and precision of measurement. Are these issues affected by multiple variables?
|
