Write a bachelor thesis: 7 tips for surveys
Many students – instead of relying on already existing statements – choose an empirical survey (or questionnaire) to complement and enrich their Bachelor thesis, which they then evaluate and analyze. Of course, these are not only exclusive, but also fresh content and insights that can greatly enhance the bachelor thesis.
Accordingly, such a survey can incorporate new perspectives, confirm or question existing theses, or raise a question for another audience in order to create a new relationship.
However, many mistakes can happen that reduce the value of the survey. The following tips help to reduce or avoid such sources of error:
Clarify your target audience.
It does not matter which department you write your bachelor thesis or which topic you are dealing with: Make sure beforehand who the target group of your survey should be. This can sometimes be simple, if you want to create a general opinion, perhaps structured only by age structure.
However, if you want to get an explicit rating from a specific group – for example, all students who have a part–time job or single parents – you should address this target group both when creating and disseminating your questionnaire.
Make your survey representative.
First, a representative survey usually looks at the number of people surveyed. After all, an objective result can not be obtained from a small number of respondents. In principle, it is therefore recommended to select the largest possible group of subjects – in the best case, several hundred participants.
Representativeness also includes who you ask and whether the participants actually belong to the defined target group. For example, if your aim is to measure employee satisfaction with their job situation, then a survey at the university, where only a small proportion are already at work, is not representative even among 300 participants.
Keep your questions clear.
During the entire time of your bachelor thesis, you will deal extensively with a topic area, acquire well–founded knowledge, are involved in the subject from early to late, and know every little thing. But please remember: the participants in your survey usually do not have this knowledge and yet still need to understand what the individual questions are about.
Therefore, keep the questions as understandable as possible. Avoid long and complicated sentences, and if you can not do without specialized vocabulary, you should definitely explain it.
Do not suggest answers.
Often one already has some idea of what result to expect in the survey – or one hopes for a certain result in order to substantiate their own argumentation and to confirm the established thesis. However, to remain scientifically objective, you must never ask leading questions in order to direct the answers in a desired direction.
In this way, you distort the result and invalidate the survey.
Keep the survey as short as possible.
It’s not always easy to find the fine line between a full survey and a much too long series of questions. When creating the questionnaire, you should always keep in mind how much you can and want to expect the participants. Just ask your fellow students how much time they would take to answer a questionnaire.
Make the answer options unique.
The question Judging on a scale of 1 to 7 … sounds always plausible, but it can bring with it the problem that the participants judge the individual values differently. While a 5 is already a very good value for a respondent, another thinks 6 and 7 are actually good.
Therefore, be specific and clarify the meaning of the answer options. For example, you could specify the following scheme: 1 point – very weak or agreeing, 2 points corresponding to a weak, 3 points to a neutral, 4 points to a strong and 5 points to a very high degree or agreement.
Start a test run in advance.
Before you start the actual survey, it may be useful to distribute a few copies of the survey to friends and family to get important feedback before you go public.
Many eyes see possible errors faster and even the first answers can help you to spot any weak points in time.