Ability grouping in secondary schools: effects on pupils' self-concepts.

BACKGROUND: Current pressure on secondary schools to increase ability grouping has raised concerns about the impact of setting on pupils' self-concepts. Evidence from previous research is conflicting. A multidimensional measure and multilevel modelling promise to clarify the effects. AIMS: This paper aims to examine the effects of structured ability grouping on year 9 pupils' self-concepts. SAMPLES: The sample comprises over 3000 year 9 pupils (aged 13-14 years) in 45 mixed secondary comprehensive schools in England. The schools represent three levels of ability grouping in the lower school (years 7 to 9). METHODS: Pupils responded to a multidimensional self-concept scale measuring academic and general facets of the self-concept. Measures of attainment were collected in English, mathematics and science. Multilevel modelling was used to examine the effect of the type of school on the general facets of the pupils' self-concept and the effects of setting in each curriculum subject on the academic facets of the self-concept. RESULTS: Pupils' general self-concept was higher in the group of schools with moderate levels of setting. The degree of setting in mathematics and science had no effect on the corresponding academic self-concepts but setting in English tended to lower the self-concepts of the higher attaining pupils and raise the self-concepts of lower attaining pupils. Gender differences were consistent with previous research, with boys having significantly higher self-concept scores than girls, except in English. These findings indicate that the degree of stratification in schools can have an impact on adolescents' self-esteem and views of themselves in school.

Explanatory Models for Relating Growth Processes.

For many purposes, longitudinal data are a great advance over cross-sectional data. The opportunities for modelling are enhanced if data for several occasions are obtained for a response, y, and at least one time-varying explanatory variable, x. The article describes, with examples, three modelling approaches when both y and x change over time. The first - a conditional approach - relates x to y in a regression framework. Earlier versions of these models were known as two-wave, two-variable (2W2V) 'causal' models. In the second, unconditional approach, growth or change parameters for x and y are themselves related in a second stage analysis. The third approach is based on structural equations modelling. All three approaches can be implemented in a multilevel framework. The article describes how multilevel models can extend the way we think about the analysis of longitudinal data, and hence how more interesting hypotheses about social processes can be modelled.

Does health-selective mobility account for socioeconomic differences in health? Evidence from England and Wales, 1971 to 1991.

The paper uses data from a sample of 1 percent of the male population of England and Wales to examine the contribution of social mobility between the censuses of 1971 and 1981 to socioeconomic differences in health. Compared to others in their social class of origin, men who had been downwardly mobile were more likely, and the upwardly mobile were less likely, to report a limiting long-term illness. However, when compared to others in their classes of destination, those who moved down reported less illness, and the upwardly mobile reported more. Prevalence of ill health in mobile men was somewhere between that in the group they left and the group they joined. Social mobility was a common event and, combined with existing socioeconomic differences in health, it acted to constrain rather than to increase these differences.

Individual differences, daily fluctuations, and developmental changes in amounts of infant waking, fussing, crying, feeding, and sleeping.

Measures of the amounts of time infants spent asleep, awake-content, feeding, fussing, and crying at 2, 6, 12, and 40 weeks of age were examined using multilevel analysis. This method enables the proportion of the variance in each behavior due to individual differences to be compared to the proportion due to age changes (development) and to day-to-day fluctuations at each age in the same infants. Day-to-day fluctuations were found to account for the largest proportion of the variance in amounts of sleeping, fussing, and crying (between 44% and 53%), testifying to the importance of instability in these behaviors as a characteristic of infancy. Against this background, both development and individual differences explained substantial proportions of the variance, with a somewhat different picture in each area of behavior. Amounts of waking and feeding were mainly accounted for by development, and no evidence of enduring individual differences was found. For sleeping, development and individual difference each contributed approximately a quarter of the variance, and the amounts infants slept remained moderately stable from 6 weeks to 9 months of age. Crying decreased linearly with age, with development accounting for 38% and individual difference 15% of the variance. Fussing proved a more stable characteristic than crying, and "high fussers" at 6 weeks of age were particularly likely to retain this characteristic at 9 months, whereas amount of crying in the first 3 months did not predict 9-month behavior. The study's clinical, conceptual, and methodological implications are discussed.

Two decades of family change: secondary analysis of continuous government surveys.

"Secondary analysis of General Household Survey and Labour Force Survey data shows how the structure of families in Great Britain has changed over the last 20 years. Dependent children are now less likely to be living in a couple family and more likely to be living with a lone mother who is either single or divorced. Families in simple households with just two generations have become more common over time. Lone mothers are now more likely to be living in simple households. The paper also considers how the number and ages of dependent children are associated with family and household type. Log-linear models are used both to smooth the data and to predict family structure in the year 2000. Gaps in our knowledge about current family structures are discussed together with implications of the findings for social policy."

Teacher expectations in infant school: associations with attainment and progress, curriculum coverage and classroom interaction.

There is still much debate, particularly in North America, about whether teachers' expectations have an effect on pupils' achievement, and through which factors expectations might be mediated. This paper reports on associations between teachers' academic expectations at the beginning, and children's attainments at the end of the school year. The study took place in infant schools in London. Associations were significant during all three years of infant school, and were not explained by children's attainments at the time of the expectation rating. Range of effects, in standard deviation units, of associations between expectations and progress over the school year ranged from 0.4 to 0.8. Two possible mediating factors between expectations and attainment were explored: differential curriculum coverage and differential classroom behaviour. It was found that children for whom teachers had higher expectations were given a wider range of activities in written language and mathematics, and this was so over and above attainments at the beginning of the school year. In contrast, there was no evidence that expectations were related to measures of classroom interaction like teacher praise and instructional contact.

Mental distress in mothers of pre-school children in inner London.

The level of mental 'distress' was assessed for a sample of mothers with pre-school children in Inner London. Fifty-two per cent of mothers had a moderate or severe distress problem in the 12 months prior to interview. Variables correlating with distress were identified and replicated in a second sample. Social class was not found to be related, and the question of the general relationship between social class and distress is considered.

Impairment of a motor skill in children with spina bifida cystica and hydrocephalus: an exploratory study.

Twenty 7-10 year old children with spina bifida cystica and hydrocephalus and 20 normals matched for age, sex and IQ were compared on a 12-trial target task, first used by Connolly, Brown & Bassett (1968). Analysis of the results, in which particular attention was paid to statistical method and to ways of analysing individual differences, showed a significant impairment in dotting speed in the spina bifida group, although both groups improved with practice. In a second experiment immediately following the first, visual monitoring of this task was restricted. The spina bifida children were initially more affected than the controls but able to recover. The findings are discussed in relation to neurological abnormalities in the spina bifida group.