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

Large-scale Assessments in Education

An IERI – International Educational Research Institute Journal

Table 1 Pooled Means and Standard Deviations of PISA Reading Literacy Scores Aggregated at the Country Level

Time points	PISA waves	M	SD	Range	N
T1	2000	474.8	53.9	322.0–549.1	42
T2	2003	482.6	41.0	374.1–543.8	40
T3	2006	466.0	51.3	311.5–556.8	53
T4	2009	470.5	43.9	370.1–540.0	56
T5	2012	475.3	45.0	384.6–544.9	56
T6	2015	473.4	42.2	359.1–527.7	54
T7	2018	470.9	40.9	371.3–525.6	54
-	Overall	473.0	45.4	311.5–556.8	355^a

The descriptive statistics were calculated based on the countries’ mean scores that form the data basis of the longitudinal analysis in the present study. For the 2015 and 2018 waves, the achievement data in PISA were represented by 10 plausible values instead of 5. Due to the longitudinal modeling, only the first 5 plausible values were considered in the present study
^aThe overall N represents the number of data points in the data set (number of countries in the sample * the respective number of repeated measurements). A total of 56 countries were included in the sample the inclusion criterion being at least 5 PISA participations between 2000 and 2018