Nick Sofroniou

Confidence Intervals for the Predictions of Logistic Regression in the Presence and Absence of a Variance- Covariance Matrix

Confidence intervals for fitted values provide valuable information about the usefulness of regression models. Although such intervals can be easily calculated using standard statistical software for response variables that have normally distributed errors (e.g., in ordinary least-squares regression), it is more difficult to calculate them for response variables that have binomially distributed errors (e.g., logistic regression). Although a number of statistical packages provide confidence intervals for fitted values directly for logistic regression models, some commonly used packages do not (e.g., SPSS). In this article we outline a method of calculating these intervals simply by fitting a model after transforming variables. This technique is evaluated by comparing results with those obtained using a method that utilizes the variance-covariance matrix. Both techniques are described in detail and applied to simple and multiple logistic regression along with step by step instructions and software commands ...

Experimental appraisal of personal beliefs in science: constraints on performance in the 9 to 14 age group.

BACKGROUND Recent curricula initiatives have promoted experimentation as a means by which relatively young children can appraise their personal beliefs and thereby modify these beliefs towards received scientific ideas. However, key psychological theories signal problems, and the enterprise is not in any event securely grounded in empirical research. AIMS As a consequence, the study reported here aimed to provide comprehensive information about childrens abilities to use experimentation to appraise their beliefs, while allowing full exploration of theorized constraints. SAMPLES The study involved 24 children at each of three age levels within the 9 to 14 range. METHODS The children were first interviewed to establish their beliefs about influences on outcome in four educationally significant topic areas: flotation, pressure, motion and shadows. Subsequently, they were asked to conduct investigations to determine whether selected beliefs were correct. RESULTS The results showed that, regardless of age or topic, very few children appreciated that to explore whether some variable is influencing outcome it is necessary to manipulate that variable experimentally and that variable only. There was a strong tendency to manipulate other variables, a tendency attributed to the intrusion of everyday reasoning practices into the experimental context. Once extraneous variables had been introduced, the children experienced great difficulties with subsequent stages in the experimental process, e.g., predicting, observing and drawing conclusions. CONCLUSIONS It is concluded that experimentation as a means of appraising beliefs is not straightforward in the 9 to 14 age group, and that the pattern of difficulties has psychological significance given the background theories. Nevertheless, while not straightforward, experimental appraisal remains possible given appropriate teacher support, and proposals are made as to the form which the support should take.

Persistence of psychological problems in adolescence: a one year follow-up study

OBJECTIVES To examine the prevalence and persistence of psychological problems in older adolescents. METHODS This study is a one year follow-up of 110 adolescents, 64 girls and 46 boys attending three secondary schools in Ireland. All were 16 at initial testing. The Youth Self Report (YSR) was the measure used. RESULTS Over a fifth of adolescents have problems in the clinical range. These problems persisted from 16 to 17. Females reported more problems than males at both ages. Some females showed a slight reduction in internalising problems at 17. Both males and females showed an increase in externalising problems at 17. Odds ratios indicate that those in the clinical ranges of the YSR at 16, had increased risk of being in the clinical range at 17 compared to those in the no problem range at 16. In comparison to those with no suicidal feeling, those with suicidal feelings at 16 were at an increased risk of still feeling suicidal at 17. More males remained psychologically healthy than girls. CONCLUSIONS A large proportion of adolescents in this sample have psychological problems and these appear to persist over a one year period. These problems impact substantially on the adolescents themselves, their families and society. Given the increasing suicide rate in young people and the persistence of suicidal feelings in this cohort, the inadequacy of mental health services in Ireland for adolescents, particularly those ages 16-18 is highlighted. The challenge of developing and providing adolescent friendly services is addressed.