Skip to main content
Large-scale Assessments in Education

An IERI – International Educational Research Institute Journal

Table 4 Discrete-time parameters for autoregression (AR) and process error for varying time-intervals derived from underlying continuous-time parameters

From: A primer on continuous-time modeling in educational research: an exemplary application of a continuous-time latent curve model with structured residuals (CT-LCM-SR) to PISA Data

Time-interval (years)

DT AR effects

95% CI for AR effects

DT process error (SD)

1

.681

[.841; .521]

6.91

2

.471

[.684; .257]

10.08

3

.329

[.546; .112]

11.53

4

.233

[.433; .033]

12.19

5

.167

[.342; − .009]

12.50

6

.121

[.271; − .029]

12.64

9

.048

[.136; − .039]

12.75

  1. CI confidence interval. A three-year interval represents the time between two PISA surveys. Values for the time-intervals 1, 2, 4 and 5 are interpolations, because there is no data for such time-intervals in the data sets
Back to article page