How were the items on the DESSA forms (First Edition) selected?

Overview

This article explains the process used to select the items on the four forms of the DESSA-mini as well as how the items and scales for the DESSA were developed. Note that this article provides information on the first edition DESSA for K-8th grades.

How were the items on the four forms of the DESSA-mini selected?

The eight items on each of the four forms of the DESSA-mini were selected from the 72 items on the DESSA using three criteria.

First, it was important for the total score on the four sets of items (the Social-Emotional Total T-score) to correlate as highly as possible to the Social-Emotional Composite (SEC) T-score on the full DESSA. This is because the primary purpose of the DESSA-mini as a screener is to predict the score that the student would receive on the full DESSA assessment. To meet this goal, items with the highest correlation with the DESSA SEC T-score were rank ordered, and the first 32 items (eight items for four forms) were chosen. 

Second, it was important for the four DESSA-mini forms to have as few items as possible while still ensuring high reliability and predictive validity; therefore eight items were selected for each form.  It is important to understand that the 8 items on each form have nothing to do with the 8 DESSA scales. Rather, these 32 items correlate the highest with the SEC T-score. 

Third, it was important for the four DESSA-mini forms to yield the same score for a given child, indicating that they are equivalent and can be directly compared.  Combinations of the 32 items were tested until the means, standard deviations, and internal reliability coefficients were sufficiently similar to determine the final composition of the four DESSA-mini forms.

How were the items and scales developed for the DESSA?

To create the items and scales for the DESSA, the authors began with a thorough literature review on resilience, social-emotional learning, and positive youth development, followed by an examination of other strength-based assessments.

Next, the readability and usefulness of the items were examined during a national pilot study.  Upon obtaining a pool of strength-based, observable items, national standardization occurred.  Utilizing the standardization dataset, the DESSA items were organized into logically derived and statistically validated scales based on the CASEL Framework (www.casel.org). The authors chose this framework because it is well-established in the research literature and is being incorporated into many state and local social and emotional learning standards.

Some minor adjustments were made, such as the inclusion of optimistic thinking as a construct.  Reliability and validity studies were then conducted.  The DESSA ended up with 72 items organized into 8 scales. For more information, please refer to the DESSA manual.