Ordination and FIC

[Gormengast]

Hi, thank you for the OpenRepGrid, it really is greatly built! I cannot seem to find in documentation these three measures, and I believe they could be useful.

1. Complexity FIC. Another useful measure of complexity is Landfield's FIC - functional independent construction (Landfield et al., 1968), which is a measure used in different studies and could be useful to compare (e.g. Space & Cromwell, 1980, Neimeyer et al., 1983)

2. Ordination. While cognitive complexity is measures the differentiation between constructs, ordination is a measure of integration, differentiation within constructs. The original should be found in (Landfield & Barr, 1976), although since this manuscript does not seem available, the description of the procedure is found in Angelillo et al., 1985.

3. Mixed self-valenced factor variance (Space & Cromwell, 1980). This measure extracts factors within which constructs can have mixed valence. Potential problem here may be how to designate positive and negative poles in the grid before the analyses are undertaken.

Also, if I can ask about indexConflict1() function. It does not produce the conflict measure? For example, in Sheehan, 1985, she operationalizes the following measures:

a. The level of ‘intensity’ was derived from the CONFLICT program and consists of the total ‘balance’ and total ‘imbalance’ values for each construct. This was used as a measure of the tightness of the construct system.
b. The ratio of ‘imbalance’ to ‘intensity’ derived from the CONFLICT program was the measure of ‘conflict’.

As I understand it, the conflict measure requires intensity measure, where one points to conflicts and the other to tightness of construing. This function therefore does not produce the above-mentioned measures? Forgive my technical ignorance if I cannot find these options.

• Landfield, A. W., Danforth, W. J., & Baugh, L. J. (1968). Functionally independent construction (FIC): Studies of consistency. Psychological Reports, 23(1), 337-338.
• Landfield, A. W., & Barr, M. A. (1976). Ordination: A new measure of concept organization. Unpublished manuscript.
• Barr, M. A. (1977). An investigation of cognitive differentiation and integration within personal construct theory. The University of Nebraska-Lincoln.
• SPACE, L. G., & Cromwell, R. L. (1980). Personal constructs among depressed patients. The Journal of Nervous and Mental Disease, 168(3), 150-158.
• Neimeyer, R. A., Klein, M. H., Gurman, A. S., & Griest, J. H. (1983) Cognitive structure and depressive symptomatology. British Journal of Cognitive Psychotherapy, 1, 65
• Angelillo, J., Cimbolic, P., Doster, J., & Chapman, J. (1985). Ordination and cognitive complexity as related to clinical depression. The Journal of nervous and mental disease, 173(9), 546-553.
• Sheehan, M. J. (1985). A personal construct study of depression. British Journal of Medical Psychology, 58(2), 119-128.

[markheckmann]

Hi @Gormengast ,

thanks for your suggestions. Would you be willing to help building them? :)

There could be various way to collaborate depending on wether or not you gave programming knowledge.

Tasks to be done are:

1. Create description of each index with explanation plus citations to include in the package help docs
2. An explanation / outline how to calculate each index with a correct calculation example (e.g. in Excel)
3. Implement a function in R that calculates the index
4. Add functions to package

I can do 3 + 4. It would be great to have help for 1 and 2 (or 3 if you know R). @Gormengast what do you think? :)

[Gormengast]

Yes, sure, I could contribute, although I don't have much programming knowledge, so I don't think I could help with 3 or 4. I can definitely help with 1. and I will try to help with 2. This of course depends on available literature and with some I may need to consult with you, since it may bring some new problems (e.g. calculating mixed self-valence requires prior designation of positive and negative poles). Thank you for wanting to implement new functions! :)

[markheckmann]

Great. I suggest to start with 2, as it is the basis for me to implement it in R. Could you create an outline of how to implement the indexes, if possible with a working example in Excel? :)

PS: "calculating mixed self-valence requires prior designation of positive and negative poles" => currently, there is no explicit way to do this, but we can just sort all positive poles on the same side as a workaround.