.2 The ALSPAC Cohort in Relation to Education Provision

The young people in the study span 3 academic years, referred to as the 'oldest', 'middle', and 'youngest' cohorts. The cases are unevenly distributed across the three years with approximately the following proportions: Oldest cohort - 25% Middle cohort - 60% Youngest cohort - 15%

2.3 Further & Higher Education, and Work Based Learning

In academic year 2007/08 the oldest cohort potentially entered Year 12 or may have left compulsory schooling. This means that they may no longer feature in the NPD, but for those undertaking further education, their educational participation should be picked up in the Individual Learner Record (ILR). In subsequent years they may also move into HE, where their participation would be picked up in the HESA Student Record. At these ages, the young people clearly become of prime interest to this department. Linking to the ILR and HESA data sets offers the potential to continue tracking the educational history of participants, offering the potential for in depth analyses of progression.

2.4 Publication Record

ALSPAC researchers have a strong track record of using linked education data. Over the past five years, 23 peer-reviewed papers have used linked data to investigate education related hypotheses.1-23 These publications used the wealth of data available from ALSPAC; incorporating the influence of genetic variation on attainment, and using detailed individual level data on social background and aspirations to help describe the impact of disadvantaged upbringings on life chances and aspirations. Evidence from ALSPAC, including linked education data, were used to contribute to the Independent Review on Poverty and Life Chances by Frank Field MP 'The Foundation Years: Preventing Poor Children becoming Poor Adults'24 and the Marmot Review 'Fair Society, Healthy Lives'.25

3. Project Summary

2.1 BIS will contract ALSPAC to conduct linkage to NPD/Data Service/HESA records to collect and process educational attainments for the ALSPAC cohort members from the age of 16 onwards. The linkage will focus on Further Education/ Vocational data provided in the ILR extract and Higher Education information provided by HESA and NPD.

2.2 A cost effective mechanism, using ALSPAC's existing linkage to the NPD, has been confirmed that will allow collection and linkage of ILR and HESA data alongside

KS5 data via the NPD. These data will follow on from the data collected from NPD under ALSPAC's contract with the DfE.

3. Proposed work

The key elements of the work to be undertaken are:

Arrange access to data & documentation Securely archive raw data Conduct linkage quality control work Anonymise the data set (remove personal identifiers, sensitive variables and replace NPD pupil IDs with a new unique ALSPAC ID) Reformat the data to ALSPAC standards to ensure compatibility with the linked schools data and self-reported participant data. Publish 'built' files of the data within the ALSPAC resource

Access and matching arrangements have been agreed as follows: Confirmation from the Department (BIS) that there is an ILR-NPD-HESA matched dataset; NPD and Dissemination Unit (Data Services Group) has advised us that the NPD-HESA matched dataset can be shared with ALSPAC under existing contractual arrangements with the Department for Education (DfE); NPD and Dissemination Unit (Data Services Group) has confirmed that the ILR-NPD matched dataset can be shared with ALSPAC once the contract with BIS has been confirmed.

Table 1 details the 5 elements of data linking included in the contract and the associated timings.

(No outline received).

(No outline received).

Analyses of data from longitudinal studies are often complicated by the presence of missing values, caused

by participant dropout or non-response. Failure to allow for this can lead to both biased and inefficient

statistical analyses. Analyses ignoring problems caused by missing data are common. Statistical research has

generated better ways to deal with these problems, but the methods are technically challenging. The

proposed research will focus on the application of multiple imputation (MI) - the most flexible available

method - in longitudinal studies. We will demonstrate the potential of MI to reduce bias and increase

precision in analyses of data from the ALSPAC birth cohort study and the ART-CC and ART-LINC HIV

cohort collaborations. We will also clarify the circumstances in which analyses allowing for missing data are

likely to have advantages over simpler methods. We will develop a framework for simulations that allow

evaluation of characteristics of imputation procedures, and use both simulations and analyses of longitudinal

data to examine how to deal with the model complexity that characterises application of these procedures.

We will adapt existing software to improve model diagnostics that may alert the user to problems in

imputation procedures and to facilitate sensitivity analyses that examine robustness of results to data that are

missing not at random MNAR). We will work with members of the CONSORT and STROBE groups, and

with journal editors, to provide guidelines on reporting analyses that deal with missing data.

(No outline received).

(No outline received).

(No outline received).

(No outline received).

Generalised Growth Mixture Models will enable us to divide the children into different groups (cluster-analysis style) and fit a different trajectory for each group. In the graph below, the y-axis is the probability of being dry at night. Preliminary analysis using data up to 91 months suggests 4 different groups: normal children, delayed dryness, severely delayed dryness and a relapse group.

Group membership can then be used as either a multinomial outcome or as a predictor for further analysis into the causes/effects of subnormal development of continence. The intention is to produce one methodological paper on the subject of fitting such models and then further epidemiological papers using the resulting groups.