For many research projects, the questionnaire is still the tool of choice. But when all the answers are finally in, confusion often sets in: You finally want to evaluate the questionnaire – and you want to do it quickly and correctly! But how do you evaluate a survey? In this article we show the 5 most important steps to evaluate a survey.
If you need assistance with your questionnaire evaluation, our statisticians will be happy to help you. Simply use our Contact form for a free consultation& a non-binding offer – or call us.
Let us know your requirements& we will provide you with a free offer within a few hours.
Google Rating
- Michael Wirth ★★★★★ one month ago
This article answers the following questions about evaluating questionnaires:
- How to evaluate a survey?
- What steps should you follow when evaluating the questionnaire?
- How to get an initial overview of the data?
- How do I detect correlations in my data??
Evaluation of questionnaires: structured procedure in 5 steps!
Of course, the exact procedure for evaluating a survey properly always depends heavily on the particular questionnaire and the research questions.
For our clients, we also offer various services for professional questionnaire evaluation: From employee satisfaction surveys to customer surveys to brand tracking. You would also like to benefit from our expertise for the evaluation of a survey? Then contact us!
In our experience, the procedure for evaluating a questionnaire can be roughly divided into 5 steps. These steps can therefore be used well as a guide when evaluating a survey.
1. Check the question
From the conception of a questionnaire and until all answers are available, some time usually passes. It is therefore useful to recall once again the actual aim of the survey. The questionnaire evaluation is then much faster and more focused.
Ideally, the problem has already been outlined and the research question developed prior to the survey. For example, a concrete research question would be: "How many people in the target group are familiar with our latest product??"If the research question has not been formulated before, it should be done now at the latest. In any case, brief reflections should be made on the following questions:
- Evaluate questionnaire: Which questions should the results answer?
- What problem do you want to solve with the results of the survey??
- Which decisions should be made on the basis of the results?
2. Prepare data and validate answers
The next step is to clean up and prepare the data from the survey. If one has used scales with several questions (e.g.B. (for customer satisfaction), this includes calculating the mean of all the questions in the scale. In the same way, the data should be checked for incorrect responses (e.g.B. Age information such as "250"). For complex data sets, you can always ask for our professional help in data preparation.
3. Analyze response rate
Only in very few cases will all the people who receive the questionnaire actually fill it out. The percentage of completed questionnaires is also referred to as the response rate. A sufficiently high response rate is crucial for the representativeness of a survey. Therefore, the response rate should be checked in every case. Is this significantly higher or lower than expected? In that case you should look for possible causes. How high a response rate should be depends strongly on the type of questionnaire. A telephone employee survey will z. B. usually have a higher response rate than a customer survey by e-mail. If data on previous questionnaires is available, a comparison to previous response rates is also useful.
In addition, it is advisable to check whether a particularly large number of participants drop out at a certain question. In this case, this question should be revised for future surveys.
4. Get an overview
The next step is to evaluate the questionnaire by getting an overview of the results. The best way to do this is to use simple descriptive statistics and illustrative visualizations. How to quickly get a first impression of the responses.
Evaluation of questionnaires: Descriptive statistics
For continuous values (scales, age, number of product purchases, etc.), it is important to have a clear overview.) statistics such as average, maximum and minimum are useful. The standard deviation can also be used to estimate how much the answers vary.
For values with groups or categories (gender, recommendation: yes/no, etc.) it is first sufficient to simply display the frequencies of the individual values. Both the absolute frequency and the relative frequency (as a percentage) should be calculated.
Evaluation of questionnaires: Visualizations
For continuous values, a histogram is recommended to show the distribution of the answers. If required, the visualization can already be divided into groups here (e.g., "group", "group", etc.). B. by gender or department).
Evaluate questionnaire: histograms are good to show the distribution for continuous values
For values with groups, both the pie chart and the bar chart have proven useful. A pie chart is useful if you are mainly interested in the relative distribution of the groups. A bar chart, on the other hand, is useful if you want to read the absolute frequency of the groups.
A simple example of a pie chart
The same data in a bar chart
5. Reveal correlations
Finally, this step is about drawing valid conclusions from the data. This step is usually the decisive one in order to answer your question and to provide guidance for future decisions. For this step, however, basic statistical knowledge is often indispensable! If you would like support for the evaluation of the questionnaire, our experts are always at your disposal. To illustrate the possibilities, we list the most common types of analysis here:
Comparison with previous results
This analysis is useful if the same questionnaire has already been used in the past. In this case, it is possible to analyze whether the new values differ significantly from the previous ones or whether different values are merely the result of random fluctuations. If sufficient data is available, trends over longer periods can also be analyzed ("Have we achieved a continuous increase in customer satisfaction over the last 5 years??").
Comparison of different groups
Are sales employees more satisfied than accounting employees? Does awareness of your product differ between people in rural and urban areas? Comparisons between groups can often reveal weaknesses or strengths of the company. But it can also be used to identify target groups that are particularly interested in your company or product.
Comparisons with benchmarks
If the Net Promoter Score (NPS) is higher than the minimum standard? How does brand awareness compare with industry standards? Comparisons with benchmarks analyze whether values from the survey meet predefined minimum standards. This can be an internally set standard or simply a comparison with the average performance in the industry.
Relationships between factors
A so-called correlation analysis can be used to determine relationships between different factors: Is the number of overtime hours related to employee satisfaction? Do customers with higher purchasing power find the design of the product more important than other customers??
With advanced analysis methods, more complex correlations can be uncovered or even predictions made: Do overtime hours only have a negative effect if the employee does not have enough autonomy in his work?? How much more is a customer from the "wealthy buyer" segment willing to pay for the better design??
Conclusion: Evaluating a survey correctly through a structured approach
So there are actually many possible answers to the question "How do you evaluate a survey?". The optimal procedure for evaluating a survey depends on the type of survey and the respective research question. However, with a structured approach, the process can usually be greatly accelerated and many mistakes can be avoided. We hope, this article could give a first overview of the evaluation questionnaire and show which possibilities there are.