The possibilities about which the hon. Gentleman expressed concern were that there could be breaches of confidentiality and that information derived from a census could be used commercially. The point that I am making is that the enumeration district is a building block for the creation of a postcode sector—not the other way round. Of course, the boundaries will not always be precisely the same. Anyone seeking to discover the social make-up of a postcode sector would not find it very difficult—give or take the odd relatively small margin of error—to build up a reasonable picture, simply by taking the information published on the basis of enumeration districts. I do not believe that the hon. Gentleman has shown that there is either a threat to confidentiality or a breach of privacy that could be exploited commercially in an undesirable way by providing information on the basis of postcode sectors.

The hon. Member for Greenock and Port Glasgow (Dr. Godman) asked me for an assurance that temporary employees would not be asked to collect information in the locality in which they lived. I can give him an absolute assurance to that effect.

The hon. Member for Leyton (Mr. Cohen) expressed concern abut the contents of a letter in last Saturday's Guardian concerning the collection of information in a longitudinal study. I refer the hon. Gentleman to a written answer that I gave to his hon. Friend the Member for Aberdeen, South (Mr. Doran), in which I set out the basis of the continuing longitudinal study that the OPCS conducts: The OPCS has no plans to conduct, commission or participate in a longitudinal survey based on the 1991 census. Such a survey would entail repeated interviewing of a sample of the population over a period of time. However, the OPCS administers a longitudinal study which can be used to link together, when required, data for a 1 per cent. sample of the population of England and Wales to produce statistical tables. The study includes the information from the various records of events held by the OPCS. The events include the 1971 and 1981 censuses, births and deaths. The study provides additional insights in the process of changes over time and increase the usefulness of information collected by the OPCS. As announced in the winter 1989 issue of Population Trends', a copy of which is in the Library, the OPCS plans to include the appropriate sample of records from the 1991 census in the study."—[Official Report, 26 February 1991; Vol. 186, c. 467.] As regards the hon. Gentleman's case that the British Computer Society study was published too late, I am bound to say that it could also have been published too early. The purpose of the study was to examine the procedures that were being put in place for the census that is to take place in April. That requires the BCS to see those procedures being put into place and it would have been absurd to advance that ahead of the time when that information would have been available to it.

The Bill is a small but significant step in completing the statutory guarantee of confidentiality surrounding the census, and I hope that the House will give it a Third Reading.