Reporting the C-score
of the Moral Competence Test*

Home | (c) Georg Lind
Last revision:
Nov. 2020

Publications on the MCT and related topics ... more

The C-score reflects a participant’s ability to rate arguments pro and contra a certain moral decision in regard to the arguments’ moral quality instead of in regard to their opinion agreement or other criteria, in short: their moral competence (see Lind, 2019).

The general reader is more interested in the participants’ moral competence rather than in a technical term like the “C-score”. The C-score is more interesting for the specialist. Please remember (and mention in your report) that the official definition of moral competence is much broader:

The ability to solve conflicts and problems on the basis of moral principles through thinking and discussion instead of through violence and deceit, and through bowing down to others. Especially, the MCT makes the respondents' ability visible to rate argument with regard to their moral quality instead of other criteria like their opinion agreement (see Lind 2019).

The C-scores ranges from 0 to 100. If your calculations resulted in negative C-scores or C-scores larger than 100, you must have made a mistake.

If you calculate C-scores for each dilemma story in order to study participants’ moral segmentation, you should get higher C-score for these dilemma stories because of mathematical reasons. If these scores are lower you must check your computer program for mistakes. You can check them easily by using the sample data on this web-site.

In a the typical report you will tell the reader the raw C-scores, e.g., C = 14.5. You do not neede to report more than one digit after the period/comma.

When you report scores for groups, you should report the mean scores. This is the best way to compare findings from different studies.

In addition, you can also report the median, especially when the distribution of the C-scores are very skewed. Some editors of journals require this. But it is not always necessary. You may also report the inter-quartile range (Q75 - Q25) and the smallest and the highest score, even if your research question does not require this. Readers may be interested to compare your findings with other findings in that respect.

In graphs, the C-score should, if possible, be always depicted on the vertical y-axis on a scale ranging from 0 and to 40, or more if higher mean scores have been found. The scale should always start from “0", so that the reader has a better chance to see the degree of moral competence in your study on a first glance. Do not spread the C-scores artificially by reducing the depicted scale range (e.g., from 10 to 20).

When reporting and displaying the C-score they do not need to be categorized as I had suggested earlier:

0,0 to 4,9 No moral competence. The reason: In typical studies with less than 20 participants in the analytical cells (e.g., men, old people, etc.) random response fluctuation might be so high that a participants with no moral competence gets a C-score higher than 0. However, if you analyze data from a study with larger cell sizes (N > 20) you may reduce the upper limit for “no competence” and interpret even C-scores of 2 and more as “a little moral competence.” If several studies find the same small C-score or small C-score difference, you can report this with more confidence. In general: The larger the sample, the freer are your data from random error.

• 5,0 to 9,9 = Some, but very low moral competence.
• 10,0 to 19,9 = Low moral competence
• 20,0 to 29,9 = Moderately high moral competence. This seems to be the threshold for preventing criminal and a-social behavior (see my compliations of experimental and correlational studies).
• 30,0 to 100 = High to very high moral competence.

Last revision: Nov. 2020

Impressum, Kontakt und Haftungsausschluss | Datenschutz | (c) Copyright von Dr. Dr. Georg Lind | KMDD® ist eine in vielen Ländern geschützte Marke