« Previous | Contents | Next »
Listen
WELL? WHAT DO YOU THINK?
ANNEX F1 - CHAID INTERPRETATION
CHAID divides a population into two or more categories that have the greatest difference with regard to the dependent variable; in this case this is the Mental Health and Vitality score or the Stigma score. It then splits each of these groups until no more statistically significant differences are found. The prime advantage of CHAID is that it repetitively searches through large numbers of linked data fields to identify the most important discriminators.
MENTAL HEALTH AND VITALITY CHAID
The Mental Health and Vitality score was derived through statements from Q2 in Section A. The 9 questions that were asked here were used to produce separate Mental Health and Vitality scores for each respondent. Both scores were scaled on to a 0-100 scale and the overall Mental Health and Vitality score was calculated using a weighted average of the 2 scores (both questions were given equal weighting i.e. 50/50%).
CHAID analysis was then used to find out which variables are driving the high and low Mental Health and Vitality scores for each person, e.g. is it demographic factors, previous experience of mental health problems etc. With the exception of the Cluster and Stigma groups, all of the breaks on the tables were put into the CHAID analysis.
The variable that had the most significant differences with regards to the Mental Health and Vitality score was Illness. This split in to 2 groups (or nodes). The first node contained those respondents who stated that they had a limiting illness. This represented 412 respondents and the average Mental Health and Vitality score for this group was 52.7, somewhat below the mean score for the total sample. The second node consisted of respondents who had no long standing illness or no limiting illness. This accounted for the remaining 947 respondents and they had a mean Mental Health and Vitality score of 70.1.
If we carry on with this node, it then splits the sample of 947 further with the variable 'Control'. This has split in to 3 groups:
Complete control. N = 158, SF-36 average = 77.9
Good deal of control. N = 572, SF-36 average = 71.0
No control, A little control, Some control, Don't know. N = 217, SF-36 average = 62.2
The first two of these nodes then split further with the variable 'Stress' which relates to the amount of stress they had experienced in the past year while the third node split on the variable 'Ease' which relates to the ease of managing on the household income.
Going back to the node consisting of respondents with a limiting illness, this split also by 'Control' and in to 2 groups.
Complete control, Good deal of control. N = 233, SF-36 average = 58.7
No control, A little control, Some control, Don't know. N = 179, SF-36 average = 43.1
The first of these nodes split again with respect to the 'Stress' variable (Completely free, Small amount vs. Moderate and Large amount) while the second split by Social Class (ABC1 v C2DE).
In terms of the group of respondents who have the highest average Mental Health and Vitality score, they can be described as follows.
There are 84 respondents within this group and they have a Mental Health and Vitality score average of 80.4
They are closely followed by the group described as:
There are 50 respondents within this group and they have a Mental Health and Vitality Score average of 79.1
In terms of the group of respondents who have the lowest average Mental Health and Vitality score, they can be described as follows.
There are 90 respondents within this group and they have a Mental Health and Vitality score average of 37.9.
« Previous | Contents | Next »