Sociometrics

Abstract

The Sociometric interview assesses peer's perceptions of children in their classroom on a variety of dimensions (Coie, Dodge, & Coppotelli, 1982). The interview is administered individually at school. Children are presented with a roster of all the students in their classroom. They first rate how much they like, or do not like, to play with each child in their classroom. On the 1st and 2nd grade forms, these ratings were made on a 3-point scale; for the 3rd grade form, this was expanded to a 5-point scale. For grade 4, no roster ratings were collected.

Individual items were scored by summing nominations across all children in the classroom. Both raw scores and z scores are available for both types of sociometric scores (unlimited and fixed). The z-scores are standardized within each classroom group. In addition, standardized social preference and social impact items were scored as follows:

  • Social Preference: standardized Liked Most - standardized Liked Least (score is then re-standardized)
  • Social Impact: standardized Liked Most + standardized Liked Least (score is then re-standardized)

Analysts should note that 13 of the z-scores were normally distributed for both the normative and the control samples. Four z-scores were positively skewed for both samples: withdrawn behavior sum z-score, prosocial sum z-score, victim behavior sum z-score, and withdrawn behavior z-score by sex. Eight z-scores were positively skewed for the normative sample but were normally distributed for the control sample: like least sum z-score top 3, like least sum z-score unlimited, aggressive sum z-score, hyperactive sum z-score, like least z-score top 3 by sex, aggressive z-score by sex, hyperactive z-score by sex, and victim behavior z-score by sex. One z-score, prosocial z-score by sex, was normally distributed for the normative sample but was positively skewed for the control sample.

One z-score, Withdrawn Behavior sum z-score, showed positive kurtosis for both the normative and the control sample. 73% of the normative sample and 77% of the control sample had scores between -1.39 and 0.0 and between -1.18 and 0.0 respectively. In addition, another z-score, Prosocial sum z-score, showed positive kurtosis for the control sample only, with 93% of the control sample scoring between -1.80 and 0.0.

Two scores, the mean roster rating given by respondent and the mean roster rating received by the respondent, were normally distributed for both samples. Another score, the standard deviation of the mean roster rating received by the respondent, was negatively skewed for the normative sample but was normally distributed for the control sample. Also, the standard deviation of the mean roster rating received by the respondent showed positive kurtosis for the normative sample with 74% of the normative sample scoring between 0.0 and 0.80.

Finally, the School Satisfaction scale was normally distributed for both the control and the normative samples. This scale showed positive kurtosis for both the control and the normative samples. 74% of the control sample and 75% of the normative sample scored between 1.0 and 5.0 (8.0 was the highest possible score for this scale). Because the internal consistency of this scale for both samples was moderate, some caution should be used when using this scale.

Dataset Names

Raw Dataset Name: CyE
Scored Dataset Name: SCMySCc

Keywords

School Adjustment, Peer Relationships, Social Preference, Status Groupings, Prosocial, Hyperactive, Victimization, Aggressive Behavior, Loneliness, Self Report.