Although our data were collected from family historians, and many of the individuals in the database formed family groups, we quite deliberately chose to focus on the analysis of individual migration and mobility over the whole life course, rather than on the movement of family groupings. This was principally because, over the life course, the household formation of an individual may change several times, and we wanted to capture the individual experience of movement over the whole life course. However, in adopting this focus we recognise that we have missed exploring other important aspects of family migration. Recent re-analysis of the data has begun to explore some of these issues (see below), but many of the questions raised by the reviewers (especially with respect to family relationships) cannot be answered from our data because the questions were never asked in our original data collection.

Because the focus of the book is on individual-level migration, I accept that we may have minimised the impact of macro-level economic and social change on migration. Clearly individual decisions about how and why people moved in the past cannot be completely divorced from larger-scale regional and national events, and we could have given more emphasis to these. However, I would still contend that for most people individual, familial and local factors were of far more importance in structuring migration decisions than macro-economic events. This is especially borne out by more detailed investigation of some of the diary and autobiographical evidence cited briefly in the book. Such personal accounts give particularly clear insights into the ways in which migration decisions were made (Pooley and Turnbull, 1997; Pooley, 1999).

The reviewers rightly raise concerns about the representativeness of the sample and, in particular, draw attention to the relative under-representation of those in low-skilled occupations. I acknowledge that this almost certainly affects the quantitative results reported in the book, probably leading to an under-estimation of the average number of moves undertaken over a lifetime, and an overestimate of the mean distances over which people moved. However, this bias was inevitable given the historical sources from which the data were drawn, and I do not believe that this seriously affects either the general conclusions of the book or the detailed interpretation of migration motives. It would, however, be very good to have more detailed studies of migration from a larger sample of those in low-skilled occupations.

It is always difficult to interpret data that measures change over time, especially when—because of source cons-traints—these data may not be directly comparable. In the book we chose to emphasise the stability of the main patterns that emerged. I accept that in doing so we may have understated the impact of change on particular localities and regions. In part this is a question of scale. The data could not be interpreted at the very local scale (where the impact of change may be most obvious) and thus the regional scale of analysis smoothed out variations and emphasised stability. It is also important to restate the fact that we were looking at gross migration flows—all moves over any distance over a lifetime—and in doing so necessarily ignored to some degree the impact of net differences in migration flows on particular localities. In this sense the conclusion of stability in gross movement that we draw does not contradict the evidence of change in net movement evidenced by aggregate census and similar sources. They are two different perspectives on the same issue and are entirely compatible with each other.

We are currently pursuing three main avenues of investigation with respect to migration and mobility—all of which relate to issues raised by the reviewers. First, we have begun to reanalyse the original database by family groupings, investigating the extent to which family relationships influenced directly the areas over which individuals moved. Although we do not have detailed information on relationships, we are able to examine the migration fields of family groups and to assess the relative importance of family links in determining these migration fields. This allows comparison with, for instance, the work of Rosental (1999), Kasakoff and Adams (2002) and Kok (2002). Provisional results suggest that most family groupings did have well defined and limited migration fields, but that the factors influencing migration were common to all migrants and were not unduly influenced by membership of a particular family grouping (Pooley and Turnbull, 2002).

Second, as suggested by Kok, we intend to return to the best of the individual life histories contained in the data base and the diary and autobiography evidence, and to seek new ways of undertaking more systematic analysis and comparison of changes in migration experience over time. We have completed a series of new quantitative analyses on the database, using logistic and ordinal regression to model migration distances, and these new analyses confirm the results presented in the book. Our next task is to focus more directly on the comparison of individual experiences at specific periods of time.

Third, we are concerned to explore in more detail the changing meaning of migration for the individual, and especially the changing relationship between residential migration and everyday mobility. It can be suggested that during the twentieth century some people have been able to substitute commuting for migration—extending their journey to work rather than moving home—thus having a significant impact on the nature and extent of residential migration (Pooley, 2002). Recent research has explored in detail the changing journey to work in Britain over the twentieth century (Pooley and Turnbull, 1999, 2000a, 2000b), and a current project is directly exploring changes in everyday mobility over the twentieth century (Turnbull, Pooley and Adams, 2002). It is suggested that future migration studies could usefully pay more attention to the interaction between residential migration and everyday mobility.

In conclusion, we are delighted that Migration and Mobility in Britain has stimulated debate and critical comment from a wide range of migration researchers. We have benefited greatly from these debates and we are indebted to all those colleagues on whose ideas we have drawn. Whilst some of the limitations of the book were inherent in the sources and methods used, both our original research and later comments have stimulated us to pursue new lines of investigation. I hope that some of these ideas may stimulate new comparative research in Europe and America so that the productive dialogue between migration researchers may continue.

Colin G. Pooley

Department of Geography

Lancaster University