• Differences Across Time

This appendix outlines the instructions to manually calculate and determine the statistical significance of differences in scores across time. To compare scores (from the same rater) across time, Reliable Change Index (RCI) values for both T-scores and raw scores of the Conners 4® and the Conners 4®–Short are provided.

Differences Across Time

Tables E.1 and E.2 provide the values needed to determine if scores on pairs of administrations are statistically different at the p < .05 level of significance for the Conners 4 and Conners 4–Short, respectively.

To use the tables:

1. Select the rater type (Parent, Teacher, Self-Report) to determine which column to use.

2. Find the row for the scale that is being compared. To be significant, the absolute difference between the two standard scores on this scale must be equal to or greater than the value in the table. A statistically significant result means that the measured change can be attributed to reliable differences between the scores, rather than random fluctuations in behavior or error in measurement. If the score has increased significantly from pre-test to post-test, then the individual’s score shows significant decline in behavior, or worsening of symptoms. If the score has decreased significantly, then the score shows significant improvement.

For example, a clinician asked a parent to rate their son using the full-length Conners 4 at Time 1 and again, a month later, at Time 2. The clinician wanted to compare the two ratings to determine whether the ratings are significantly different from each other. The parent’s rating on the full-length form resulted in a Time 1 Emotional Dysregulation T-score of 80 and a Time 2 T-score of 70. Using a significance level of p < .05, this 10-point difference between Time 1 and Time 2 is greater than the value of 7 shown in Table E.1, indicating statistically significant improvement. Conversely, the parent’s ratings on the Hyperactivity Scale (Time 1 = 74, Time 2 = 70) did not change significantly, because the 4-point difference between scores is smaller than the value of 9 shown in Table E.1. These results for Hyperactivity suggest that continued or different intervention may be needed.

In most cases, the minimum values needed for significance should be used when comparing T-scores. Using the minimum values needed for significance when comparing raw scores is appropriate when a youth changed norm groups across administrations (i.e., had a birthday). Raw score critical values can be found in Tables E.3 and E.4 for the full-length Conners 4 and Conners 4–Short, respectively.