TEDS Data Dictionary

8 Year Study

Contents of this page:

Introduction

This study is labelled the "8 Year" study for convenience, although in fact twin ages varied from 7 to 9 years when the data were collected. It has also been referred to as the "CAST" study; however, the questionnaire contains other measures as well as the CAST. Furthermore, the CAST measure has been re-used in later TEDS studies.

The 8 Year study data were collected by means of a single short parent questionnaire. The items in the questionnaire are twin-specific. The measures used in the questionnaire are described in full in a separate page.

The 8 Year data were collected between 2002 and 2004 from the entire TEDS sample (twins born between 1994 and 1996). Data collection was not precisely timed to coincide with a particular twin age. Twin ages varied from roughly 7 to 9 years, with an average around 8 years, when the data were collected.

The sample

The study sample included all TEDS families except for the following: (a) those that had withdrawn from TEDS; (b) those with address problems, who could not be traced; and (c) families in which either twin had a severe medical condition other than ASD/Autism (the latter were of particular interest in this study).

The 8 Year questionnaire was sent to roughly 13950 of the 16810 families in the original TEDS sample from ONS. Hence there were around 2900 families that were not sent the 8 Year questionnaires. Of these, roughly 1150 had withdrawn, roughly 1400 were known address problems, and around 350 were excluded on medical grounds.

The data returns for the 8 Year study are summarised in a separate page. There are further pages comparing samples and returns for different TEDS studies.

Data collection

Each family in the sample was sent a questionnaire (pdf), incorporating a consent form on the front page. The measures used in the questionnaire are described in detail on another page. Note that all data were parent-reported.

Questionnaires were mailed out and collected in two 'waves', summarized in the table below:

Wave: Wave 1 Wave 2
Cohort: 1 2 3 4
Twin birth dates: Jan-94 to Aug-94 Sep-94 to Aug-95 Sep-95 to Aug-96 Sep-96 to Dec-96
Mailing date: November 2002 October 2003
Approx. twin ages when data returned: 8.2 to 9 years 7.2 to 8.5 years 6.9 to 7.5 years
Other materials mailed: Sent simultaneously with the 9 Year Study consent form. Sent without any other study materials. Sent simultaneously with the 7 Year Study parent booklet.
Who did the mailing: Group Sigma (except for overseas and autistic families, who were mailed by TEDS) TEDS

As indicated in the table above, for convenience it was decided to mail the 8 Year questionnaire simultaneously with 7 Year and 9 Year study materials for some cohorts. For most families, the mailing of the questionnaires was contracted out to Group Sigma. However, some families were mailed directly from the TEDS office: all overseas families; all families known to have autistic twins; and all families in the final cohort (organised alongside the 7 Year mailing).

In wave 1, the 8 Year questionnaires were returned directly to Group Sigma, and were only returned to TEDS after scanning. Group Sigma were asked to log the questionnaires as they were received, and to send regular updates to TEDS to enable us to keep our records up to date. This proved problematic, so for wave 2 the questionnaires were returned directly to TEDS instead; they were then sent to Group Sigma for scanning at a later date.

Up to 3 written reminders were sent to families who did not return the questionnaires promptly.

Data entry

General data entry issues (for all studies) are described in a separate page. Data from the 8 Year questionnaire were taken up electronically by optical scanning. The layout and design of the questionnaire had been designed with this in mind. This was handled externally by Group Sigma, a commercial company (as mentioned above, they also handled most of the mailing).

The questionnaires were delivered to Group Sigma from the TEDS office in batches. After scanning each batch, Group Sigma returned the data in plain text files (the data files are described in detail in another page).

The questionnaire contains a few free text fields (in the consent form on the front page, and in one question on the final page). However, no attempt was made to take up these fields electronically, other than a data flag to indicate whether anything had been written in the signature box at the bottom of the consent form. Thus, the electronic data from the questionnaire are entirely numeric. There is an annotated version of the questionnaire (pdf). This document shows, in green, the scan positions used for each field in the Group Sigma raw data files. It also shows, in blue, the field names and value codes that have been used in the cleaned raw data.

On the front page of each questionnaire, the FamilyID, parent name, twin names and twin birth orders were printed. Also on the front page, there is a tick box labelled "please cross this box if the birth order is incorrect". The intention here was to identify families where the twin birth order recorded in TEDS (and printed on the questionnaire) was incorrect. However, a surprisingly large number of parents ticked this box in the returned questionnaire. Some of these parents were contacted by telephone for confirmation, and it was found that (in most cases), the parents had misunderstood this instruction. The data from this tick box are therefore of dubious value.

The primary electronic raw data, as returned by Group Sigma, were returned in plain text files, which have been retained for reference. After data cleaning, the data were aggregated and stored in a single Access database file. This Access database is now treated as the master copy or source of all 8 Year data to be used in analysis. All the raw data files are described in full in 8 Year raw data files.