Using Autoregressive Moving Average Model to Investigate the Stability of Dimensions of Emotional Support
| dc.contributor.advisor | Curby, Timothy W. | |
| dc.contributor.author | Hamdan, Noora | |
| dc.creator | Hamdan, Noora | |
| dc.date | 2013-07-02 | |
| dc.date.accessioned | 2013-09-09T17:10:24Z | |
| dc.date.available | 2013-09-09T17:10:24Z | |
| dc.date.issued | 2013-09-09 | |
| dc.description.abstract | This study investigated the minute-to-minute stability of dimensions of teacher Emotional Support in pre-kindergarten classrooms. Developmental theory states that the proximal processes, in this case, the teachers’ Emotional Supportive interactions, occur in the moment-to-moment interactions between a person and their environment and drive development. The present study examined the stability of those interactions, on the time scale in which they take place, to appropriately align our analytical methods with our theory of development. That is, how stable are minute-to-minute classroom Emotional Supports over time? Furthermore, when conducting observations of a lesson, raters may increasingly determine that a score on a dimension for a teacher fits as opposed to actually making independent ratings at each time point. In this way, ratings of teachers’ Emotional Support may become more stable over time as raters increasingly make up their minds. Then the question is not only how stable are the ratings but also, to what degree are ratings stabilizing over a lesson? In combination, we can better understand the experiences of children during a lesson. Participants were randomly selected from publically-funded pre-kindergarten programs where the majority of children were eligible to enter kindergarten the following school year. Data were coded from videotapes of teachers using an adaptation of the Classroom Assessment Scoring System (CLASS) (Pianta, La Paro, & Hamre, 2008), whereby dimensions of Emotional Support (positive climate, negative climate, teacher sensitivity, and regard for student perspectives) were coded once every minute during a language arts lesson. To address the research questions, autoregressive moving average (ARMA) models were fit to each dimension of Emotional Support. ARMA modeling is based on the notion that repeated measurements are correlated across time and may be expressed as an autocorrelation function. The autoregressive portion of the model answers the question of how stable Emotional Support dimensions are. The correlations between error variances were examined to determine if the correlations are negative which suggests that raters were increasingly making up their minds about a given rating. Findings showed that overall Emotional Support and all individual Emotional Support dimensions showed low levels of stability. Results showed that raters are becoming more consistent in their ratings of Average Emotional Support and its dimensions over time. These findings allow for a better understanding of rater effects in classroom observations. With the recent emphasis by federal agencies on the use of observation, understanding the extent of rater effects has large policy implications. | |
| dc.identifier.uri | https://hdl.handle.net/1920/8431 | |
| dc.language.iso | en | |
| dc.subject | Autoregressive | |
| dc.subject | Moving average | |
| dc.subject | Teacher emotional support | |
| dc.subject | Stability | |
| dc.subject | Rater effects | |
| dc.title | Using Autoregressive Moving Average Model to Investigate the Stability of Dimensions of Emotional Support | |
| dc.type | Thesis | |
| thesis.degree.discipline | Psychology | |
| thesis.degree.grantor | George Mason University | |
| thesis.degree.level | Master's | |
| thesis.degree.name | Master of Arts in Psychology |