M3.07. Conducting evaluation of the mentoring process

4. Analysing and Interpreting Data

4.1. Quantitative Data Analysis and Interpretation of Results

Descriptive statistical analysis
When conducting quantitative analysis, you will need some basic statistical analysis skills.
When you calculate the number and percentage of responses to a particular question or calculate the average rating for questions about the usefulness of the training, you are starting to do descriptive statistical analysis. It is used to examine the responses to a question by calculating and looking at the following things:

  • Distribution of responses or frequency distribution (e.g., how many people checked response option 1, response option 2, response option 3, etc.).
  • Average value, or the mean (i.e., looking at the average rating across the participants' ratings).
  • The most common response, or the mode.
  • The number in the exact middle of the data set, or the median.

Descriptive statistical analysis also provides another piece of information technically referred to as variability, which refers to the following:

  • The spread of your results, including the range (difference between the highest and lowest scores).
  • Variance (shows how widely individuals in a group vary in their responses).
  • Standard deviation (how close or far a particular response is from the average response).

Quantitative findings must be interpreted in the context of the organisation/programme and these questions can guide your interpretation:

  • What is the scope of the programme's impact, or how effective was the programme?
  • If you had a lot of missing data or a poor response rate, why and what can be done differently to increase the response rate in the future?
  • Are the results what you expected when you planned the programme? If not, what do you think affected the results? Do you have qualitative data that can provide further insight into the results?
  • Are the results significant to the mentoring programme or not, regardless of statistical significance, and what does it mean? For instance, the difference in responses from two groups of people might not be statistically significant but could still be large enough to motivate change within the programme activities.
  • What implications do the results have on the programme? What actions do you need to take, if any?