Keskiviikko 28.8.2019
klo 12.30 – 14.15 (huomaa session pidennetty aika)
Sali: A3 (Päätalo)
Puheenjohtaja: Minna Torppa, Akatemiatutkija, Opettajankoulutuslaitos (Jyväskylän yliopisto
Tässä sessiossa esitellään viisi tutkimusta, joissa tutkijat esittelevät tuoreita tutkimuksiaan lasten ja nuorten kehityksen alueella. Tutkimuksissa on käytetty erilaisia pitkittäisaineistoja ja erilaisia menetelmiä vastaamaan tutkimuskysymyksiin. Menetelmät pohjautuvat rakenneyhtälömallinnukseen.
Minna Torppa esittelee mallia, jossa tutkittiin lukutaidon ja lukuinnostuksen kehityksen yhteyksiä luokilla 1-9 (random intercept cross-lagged panel model). Noona Kiuru esittelee kiusaamisen ja käyttäytymisen ongelmien kehityksellisiä alaryhmiä (mixture modeling). Eija Pakarinen esittelee tutkimusta joka tarkasteli oppilaiden ja opettajien vuorovaikutuksen yhteyttä lapsen tehtävästrategioihin (multilevel cross-lagged panel model). Tuomo Virtanen esittelee tutkimustaan, joka tarkastelee muutosta kouluun sitoutumisessa (second-order latent curve growth models). Kati Vasalampi esittelee sitoutumisen tuntikohtaista muutosta tarkastelevaa analyysiä (SEM & BAYES estimation). Alla esitellään kukin tutkimus yksityiskohtaisemmin. Esitykset ovat suomeksi.
Minna Torppa (Akatemiatutkija, Opettajankoulutuslaitos, Jyväskylän yliopisto)
Pekka Niemi, Kati Vasalampi, Anna-Maija Poikkeus, & Marja-Kristiina Lerkkanen
Leisure reading (but not any kind) and reading comprehension support each other – A longitudinal study across Grades 1 and 9
This study examines associations between leisure reading and reading skills in a data of 2,525 students followed from age 7 to 16. As a step further from traditional cross-lagged analysis, a random intercept cross-lagged panel model (RI-CLPM) was used to identify within-person associations of leisure reading (books, magazines, newspapers, and digital reading), reading fluency, and reading comprehension. The modeling strategy chosen follows from the concern that the traditional cross-lagged model does not yield interpretable estimates due to the mixing of between-person variance (stable differences between individuals across time) and within-person variance (fluctuations around the stable level at each time-point; e.g., Berry & Willoughby, 2017; Curran et al., 2014; Hamaker et al., 2015). In this study we applied RI-CLPM models based on as suggested by Hamaker et al. (2015). We make, however, an important addition to the model by using latent factors in order to handle measurement error in line with a model recently utilized by Seppälä et al. (2015). The inclusion of several indicators for reading fluency and leisure reading allowed us to build latent factors and thus omit measurement error. Result showed significant between-level correlations but also various within-level cross-lagged associations. In Grades 1 to 3 poorer comprehension and fluency predicted less leisure reading. In later grades more frequent leisure reading, particularly of books, predicted better reading comprehension. Negative associations were found between digital reading and reading skills. The findings specify earlier findings of correlations between individuals by showing that reading comprehension improvement, in particular, is predicted by within-individual increases in book reading.
Tiina Turunen, Noona Kiuru, Elisa Poskiparta, Pekka Niemi, & Jari-Erik Nurmi
Word reading skills and externalizing and internalizing problems from Grade 1 to Grade 2: Developmental trajectories and involvement in school bullying in Grade 3.
School bullying is associated with externalizing and internalizing problems, but little is known about whether reading difficulties also play a part. We asked how, in Grades 1 and 2, word reading skills and externalizing/internalizing problems predict the degree to which students are involved in bullying in Grade 3. Using a sample of 480 Finnish children (Mage = 7 years2 months at the beginning of the study), developmental profiles were identified using mixture modeling based on reading skills, as well as externalizing and internalizing problems. In Grade 3, one fifth of the students were involved in bullying as victims, bullies, or bully/victims. Poor readers with externalizing/internalizing problems were most involved as bullies and bully/victims but not as victims. Average readers with externalizing/internalizing problems were also involved in bullying, whereas students with only reading difficulties were not. Skilled readers displayed little externalizing/internalizing problems and were not involved in bullying.
Eija Pakarinen (apulaisprofessori, Opettajankoulutuslaitos, Jyväskylän yliopisto)
Marja-Kristiina Lerkkanen, & Antje von Suchodoletz
The quality of teacher-child interactions in relation to task-focused behaviors
Teacher-child interactions in the classroom are among the most important factors in promoting children’s social-emotional and academic outcomes. However, less is known about to what extent the quality of teacher-child interactions is associated with children’s task-related behaviors. The present study aimed to investigate the cross-lagged associations between the quality of teacher-child interactions and children’s task-focused behaviors. The sample comprised 49 kindergarten teachers and 495 children (51% boys) who participated in a larger study (Teacher Stress Study; TESS). The data were collected twice across the kindergarten year (fall and spring). At each wave, teacher-child interactions were video-recorded during one school day and rated using the Classroom Assessment Scoring System (CLASS Pre-K; Pianta et al., 2008). In addition, teachers rated children’s task-focused vs task-avoidant behaviors in learning situations using the Behavioral Strategy Rating Scale (BSRS; Aunola et al., 2000). There were significant differences between classrooms in children’s task-focused behaviors (ICCs = .07 and .12 in fall and spring, respectively). Cross-lagged multilevel path models were specified to investigate the links between quality of teacher-child interactions and task-focused behaviors. Separate path models were specified for each of the CLASS domains (emotional support, classroom organization, and instructional support). Teacher work experience, class size, child’s gender, number sequences skills, and parental vocational education were controlled for in the analyses. The results indicated that high-quality emotional support was related to more subsequent task-focused behaviors typical of the classroom. There was a bidirectional link between task-focused behavior and observed quality of classroom organization: task-focused behavior typical of the classroom was related to higher subsequent quality of classroom organization and classroom organization increased subsequent task-focused behaviors in a classroom. The results emphasize the importance of high-quality teacher-child interactions for promoting children’s school engagement and task-focused behaviors.
Tuomo Virtanen, Yliopistonlehtori, Kasvatustieteenlaitos, Jyväskylän yliopisto, Marja-Kristiina Lerkkanen, & Kati Vasalampi
Development of student engagement from primary to lower secondary school: Associations with disengagement from upper secondary school.
Using the First Steps Study data set consisting of approximately 1,800 Finnish students, the present study examined the longitudinal associations between student engagement and disengagement. Students were surveyed four times: at the end of primary school (Grade 6), at the beginning of lower secondary school (Grade 7), at the end of lower secondary school (Grade 9), and in the first year in upper secondary education. The study had the following aims. First, we examined the extent the initial level and the rate of change of student engagement (exerting effort on schoolwork forming behavioral engagement and experiencing support from teachers, peers, and family forming affective engagement) from Grade 6 to Grade 9 predict disengagement (truancy and intentions to quit) from upper secondary school. Second, we examined the extent students’ affective disengagement (cynicism about the value of school) associates with concurrent levels of student engagement at Grades 6, 7, and 9. We used second-order latent curve growth models (SGMs: Geiser, Keller, & Lockhart, 2013; Wickrama, Lee, O’Neal, & Lorenz, 2016) as the modeling strategy. SGMs is an optimal method to examine multidimensional constructs such as student engagement because it enables the use of multiple repeatedly measured observed variables. Other advantages of SGMs include separation of the measurement error from true change, ability to test longitudinal factorial invariance, greater statistical power to detect individual differences in change, and isolation of indicator-specific effects from shared construct variance. Moreover, unlike first order latent growth curve models, SGMs do not ignore the presence of time-specific variance but separates it from random error, thus providing more accurate reliability estimates (Geiser & al., 2013). This is an important benefit given that student engagement can be assumed prone to time-specific influences (unlike for example, intelligence scores). The results showed that higher initial level at Grade 6 and increases in student engagement from Grade 6 to 9 predicted less truancy in upper secondary education. The higher the initial level of student engagement, the less upper secondary school students reported intentions to quit. Cynicism about the value of school correlated negatively with concurrent levels of student engagement at Grades 6, 7, and 9.
Kati Vasalampi, yliopistotutkija, Psykologian laitos, Jyväskylän yliopisto, Joona Muotka, Marja-Kristiina-Lerkkanen ja Kaisa Aunola
Intra-individual Dynamics of Engagement Experiences: Transfer Effect from Previous to the Following Lesson
In the present study, we examined intra-individual transmission of students’ experiences of engagement, that is, behavioral and emotional engagement, disaffection and help-seeking, from one lesson to the following lesson. The sample consisted of 57 Finnish lower secondary school students attending Grade 7. The students filled in mobile ratings on their lesson-specific engagement experiences at the end of each lesson during one week (5 days). Each student rated, an average, 17.1 lessons (SD = 5.67). The analyses were carried out using dynamic structural equation modeling and the parameters of the models were estimated using BAYES estimation. In the multilevel analyses a statistical model with two levels was specified: at the within-level we estimated variation within individuals (i.e., between lessons) and at the between-level we estimated variation between individuals. We examined whether students’ engagement in a lesson (yt+1) was predicted (a regression coefficient β) by engagement in the previous lesson (yt). To do so we specified a lagged variable of lesson-specific engagement (yt+1) based on concurrent lesson-specific engagement (yt). In this context, the data was reorganized so that the first lesson of each day was not predicted by the last lesson of the previous day by setting (yt+1) variable to be missing in the last lesson of the day. The time-points within each day were assumed to the equidistant. Then, models with random slopes were utilized. First, we examined transmission in each engagement dimension separately (lagged effects). Second, cross-lagged paths from one engagement dimension to another were added to the previous model to examine lagged relations between all factors simultaneously. The results indicated statistically significant intra-individual transfer effect in students’ classroom experiences of behavioral and emotional engagement, disaffection and help-seeking from previous to the following lesson regardless of subject content. High behavioral or emotional engagement in a particular lesson predicted also low level of disaffection in the following lesson when the level of disaffection during previous lesson was controlled for. One found intra-individual transfer effects varied between individuals: Transmission of behavioral engagement from lesson to lesson was stronger among girls than among boys. The present study provides important information to teachers by showing that the student’s experiences of each lesson can carry over to the next lesson although the study context changes.