The education systems of the 21^st century, aiming to enhance also thinking, benefit of tools for assessing thinking and learning to learn competencies. Large-scale assessments on computer-based platforms gives new possibilities for evaluating students' effort in these assessments by utilizing log data.

Diagnosis and evaluation in the learning process has proved its usefulness in the development and improvement of the education system. The longitudinal assessment of students’ learning is a power of pedagogical purposes and which can be used for diagnosing what is needed to move forward in the development of responsibility for learning. (Deakin Crick et al., 2004)

Learning to learn is an important educational goal. Also in the literature there is no general agreement about the definition of ‘learning to learn’ (Candy, 1990) ‘Learning to learn’ could be described as “a skill, or more plausibly as a package of skills, involving study skills, critical analysis, time management, planning, goal setting and so on” (Rawson, 2000, p. 225).

The Finnish learning to learn test as preparedness for learning consists of cognitive tasks measuring reading comprehension, mathematical thinking skills and reasoning skills, and questionnaires measuring learning-related attitudes (see Hautamäki et al., 2002). The attitude scales used in this study come from the theoretical backgrounds of achievement goal theory (e.g., Elliot & Dweck, 1988; Harackiewicz et al., 2002), and agency beliefs theory (e.g., Chapman, Skinner, & Baltes, 1990).

 As many studies show, when entering school, many pupils show a different level of cognitive development as well as certain skills. These levels can change in the learning process. Support, diagnosis and evaluation in the learning process has proved its usefulness in the development and improvement of the education system.

This paper reports on a longitudinal study in which the development of pupils’ preparedness for learning has been followed throughout the nine-year comprehensive school. Longitudinal assessments are useful and even necessary in order to understand the mechanisms of the development of performance.

The aim of the present study is to analyze how the four (using 25 % grouping) performance group levels in assessment of cross-curricular learning to learn skills change from 1^st to the end of the 9^th grade comprehensive school.

The specific research questions are:

1. What is the stability and trends within the 1^st grade performance level classification of pupils into four groups, when students are followed thru the 4th, 6th and 9th grades?
2. Are there systematic group level effects in explaining the variation of cross-curricular learning to learn skills?  

Learning-to-learn skills was administered as a group test to 40 first grade classes (N=744) in autumn 2007. It worked as a baseline measurement for a 9-year longitudinal study on learning-to-learn, to which the 16 participating schools were randomly selected from the schools in the capital area using an equal-probability method that ensured representativeness with regard to socio-economic status. The Learning-to-learn skills test at the 1st grade comprises six cognitive tasks, and two drawing tasks, which will not be discussed in this paper. Two cognitive tasks are versions of classical Piagetian tests for the development of horizontal and vertical axes (Piaget & Inhelder, 1956; Hautamäki, 1984). The visuo-spatial memory task is adapted from Wilson & al. (1987; modified by Logie and Pearson, 1997), and the auditory memory task has been developed for this screening test. Analogical reasoning skills are measured by a task adapted from a Dutch geometric analogies test (Hosenfeld & al., 1997). Pupils’ capacity to follow teacher’s instructions is assessed by a task developed for this screening test, in which the pupils draw a path on an empty grid following teacher’s dictation. Learning-to-learn skills were measured twice ( 4-6th grades=, and school achievement records were measured by teacher-reports. Teachers and parents completed questionnaires of each pupil’s learning, behavior and social skills. The students changed their schools due to different reasons, and they were followed in their new schools leading to that the 7th grade sample was extended to the new learning environments, to include 29 schools and 2500 students. Furthermore, the 9th grade measurement was extended to cover all the municipal schools: 55 schools and the number of students close to 4000. It is important to notice that in the analysis only the pupils that were part of the study from the very beginning in the first grade were taken into account. The 1st grade grouping of pupils into four 25 % groups was based on the average of the cognitive scales. The follow-up of students in their groups was done with repeated measures ANOVA. We will use also the structural equation models (SEM).

Preliminary studies have shown that there are (structural, i.e, correlational) connections between learning to learn test as preparedness for learning results in the beginning of the elementary school and their performance on standardized mathematics and reading tests at the end of the second grade (Vainikainen, 2011). Our results first indicate that the four different groups of preparedness for learning follow different types of development during the school in 4-6-th and 9th grade. The dynamics of the development of Learning-to-learn skills has changed and all groups are approaching more average values, which was more noticeable for the weaker group. However, at the same, the best group lost its clear positions and the level of cognitive performance of students in this group decreased at grade 6. It is of interest to note that the general trend in all of these groups was toward average mean level, implying also the decrease in variation. The changes within groups and individually will be modelled using SEMs to analyse the roles of gender and educational background. The relations of learning-to-learn competencies and group-membership will be analysed also in relations to school marks and teacher- and parent-reported social skills.

Candy, P. C. (1990). How people learn to learn. In R. M. Smith et al., Learning to learn across the life span (pp. 30–63). San Francisco: Jossey-Bass. Chapman, M., Skinner, E. A., & Baltes, P. B. (1990). Interpreting correlations between children's perceived control and cognitive performance: Control, agency or means-ends beliefs. Developmental Psychology, 26 (2), 246–253. Deakin Crick. Learning how to learn: the dynamic assessment of learning power. The Curriculum Journal Vol. 18, No. 2, June 2007, pp. 135 – 153. Demetriou, A., & Kazi, S. (2006). Self-awareness in g (with processing efficiency and reasoning). Intelligence, 34, 297-317. Elliot, A. J., & Dweck, C. S. (1988). Goals: an approach to motivation and achievement. Journal of Personality and Social Psychology, 54 (1), 5–12. Harackiewicz, J. M., Barron, K. E., Pintrich, P. R., Elliot, A. J., & Thrash, T. M. (2002). Revision of achievement goal theory: necessary and illuminating. Journal of Educational Psychology, 94 (3), 638–645. doi: 10.1037//0022-0663.94.3.638 Hanushek, E.A., Kain, J.F., Markman, J.M., & Rivkin, S.G. (2003). Does peer ability affect student achievement? Journal of Applied Econometrics, 18, 527-544. Hautamäki J., Arinen P., Eronen S., Hautamäki A., Kupiainen S., Lindblom B., Niemivirta M., Pakaslahti L., Rantanen P., & Scheinin P. (2002). Assessing Learning-to-Learn. A Framework. National Board of Education, Evaluation 4/2002. Rawson, M. (2000). Learning to learn: more than a skill set. Studies in Higher Education, 25 (2), 225–238. 322 K. Waeytens et al. / Learning and Instruction 12 (2002) 305–322 Schnipke, D. L. & Scrams, D. J. (2002). Exploring issues of examinee behavior: Insights gained from response-time analyses. In C. N. Mills, M. T. Potenza, J. J. Fremer, & W. C. Ward (Eds.), Computer-based testing: Building the foundation for future assessments (pp. 237–266). Mahwah, NJ: Erlbaum. Sheppard, C., & Gilbert, J. (1991). Course design, teaching method and student epistemology. Higher Ed). Response time effort ucation, 22, 229–249. Song, H.-D., & Grabowski, B. (2006). Stimulating Instrict Motivation for Problem Solving Using Goal-Oriented Context and Peer Group Composition. Educational Technology Research and Development, 54 (5), 445–466. Vainikainen,M.-P, Gustavson, N., Kupiainen, S.,& Marjanen, J.(2011). Validity of a Learning-To-Learn Screening Test for First Graders in Predicting Later Success in Mathematics and Reading. Berlin: ECER-2011 Vainikainen, M.‐P. (2014). Finnish primary school pupils’ performance in learning to learn assessment: A longitudinal perspective on educational equity. University of Helsinki, Department of Teacher Education Research Report 360. Helsinki: Unigrafia