How to prepare for a consultation
For an effective consultation, you may think about the following questions.
When should the consultation take place?
The best time to start a consultation is in the planning phase of your project. Here, for example, questions about data collection can be clarified, which simplify the later evaluation or make it possible in the first place.
What should you bring to the consultation?
Think about your objectives and ask yourself:
- Why you are undertaking a consultation and what it is you want to find out?
- Which questions do you primarily want to analyze, which are only secondary questions?
- What are the dependent and independent variables?
- What is the format of the data?
- How was the data collected? How should these be collected?
Of course, a consultation is always individuell. In particular during consultation in the planning phase, you should be able to provide information on the structure of the test, measuring ranges, maximum sample size or maximum costs. Also of interest is the accuracy in which your question should be answered. If you already have data or results from previous examinations, these can also be helpful in planning the examination.
In addition to the data itself, in most cases study plans, protocols or a code plan are also important.
How should I prepare the record?
- For computer-aided analysis of the data, the data must be available in suitable formatting. For example, if you have JMP or SPSS, we recommend that you use them to enter and save the data. Alternatively, you can save the data in a form used by Microsoft Excel (* .xls) or in ASCII format (* .txt) with tab delimitation.
- Each feature (each variable) must be listed in a separate column and given a name. The name of the feature is in the header or in the first row of the column and the observations in the following rows.
- You can not use more than one line for the description of the characteristics.
- Different features or repeated measurements are to be identified by different names.
- Please do not write any comments, explanations or own calculations in the table itself.
- The same characteristic values are always to be coded the same. The use of characters or strings is recommended for encoding nominal features.
- If there are missing values, leave the corresponding cells (fields) empty (no entry).
- Spaces are to be avoided.
An example can be found here:
