Evaluation Methods in Empirical Research

After data has been collected, it must be appropriately analyzed and evaluated. 
Similarly to the other chapters, the evaluation of data must also be examined in a differentiated manner. Here too, a distinction must be made regarding how the data was collected. Was it obtained using quantitative or qualitative research methods? Depending on this, different approaches must be taken. 

Evaluation Methods of Quantitative Research

To analyze and evaluate quantitative data, there are two fundamental types of statistics: 

• Descriptive Statistics 
• Inferential Statistics 

In descriptive statistics, values and quantities of results are summarized "quite simply". Frequency distributions are created in diagrams, means and standard deviations are calculated, and correlations are determined. The following calculation methods are crucial in this area of statistics:

• Measures of Central Tendency: Arithmetic mean, median, and mode
• Dispersion: Range (difference between highest and lowest value), standard deviation and variance, interquartile range 
• Correlation: The correlation coefficient indicates the relationship between two variables and ranges from -1 (negative relationship) to +1 (positive relationship). A value of 0 would mean that no statistical relationship could be determined. In other words, the variables examined have no proven relationship.

In inferential statistics, the question is whether the results were merely coincidental or whether scientific laws apply. The most important tool in inferential statistics is hypothesis testing:

• Hypothesis Testing: Hypothesis testing is used to determine whether conclusions can be drawn from the sample to the entire population. For this purpose, an alternative hypothesis (H1) is set against the currently valid hypothesis (null hypothesis) in order to ultimately determine whether it can be confirmed or refuted. At the end of the investigation, a researcher can state that with a margin of error of X percent, the hypothesis can be confirmed or refuted. 

Inferential statistics therefore represents an important part of quantitative evaluation. It classifies the descriptive results and substantiates the conclusions in your work.

 

Evaluation Methods of Qualitative Research

The challenge in qualitative research is to be able to analyze the collected data in some way, as it must first be documented and compiled through, for example, interviews.
Therefore, the following process must be followed in qualitative research:

1. Recording of Data: With the help of recordings or documentation, interviews can be preserved
2. Preparation (Transcription) of Data: You transcribe each interview word for word in a text document and thus document very precisely what can be heard in the recordings. Here, however, it is solely about the contents of the respondent that are important to the topic. Dialects, speaking pauses, and filler words (e.g., "um") can be omitted. 
3. Evaluation Method: The documented data can now be analyzed. There is not just one approach to analyzing qualitative data. Possible methods include: qualitative content analysis, grounded theory, typological analysis, object-related theory building, sequential analysis, etc. 

For simplification, we show here the procedure of summary content analysis according to Mayring (a form of qualitative content analysis):

Phase 1: Paraphrasing - Uniform language level, remove repetitions and elaborations.
Phase 2: Generalization - The paraphrases must be brought to a level of abstraction. You generalize the paraphrases so that they become more general.
Phase 3: Reduction - Only phrases that are important are retained. Duplicates or unimportant ones are deleted. And paraphrases that are the same or similar are summarized.

Example:
Phase 1: "Personal contact is essential" and "Trust can only be built through personal contact"
Phase 2: "Personal contact is indispensable" and "Trust is only possible through personal contact"
Phase 3: Personal contact is important because it makes trust possible

In this way, at the end you have compressed various statements into a category system. Now you must still verify that all statements from the first phase are included; otherwise you would need to restart the process. You can then use this summary in the further course of the investigation to interpret it and draw conclusions for your research question. 

 

And Now? Time to Write Your Report

You have now successfully evaluated the collected data and analyzed it with various tools. Now it's about interpreting the insights you've gained, placing them in the current context of research, and documenting the results in a report. This happens in an academic paper. How you write it, how it's structured, and what you need to keep in mind, we'll show you in the next chapter.

 

---

 

You will learn in the next chapter:

• The Proper Structure of an Academic Paper