| pubmed-article:1625763 | pubmed:abstractText | Large scale surveys have conventionally been analysed using a statistical package. This saves considerable time and permits many more analyses to be undertaken than with more traditional methods. It does, however, have certain drawbacks. First, it does not help with the problem of data management. SPSS and Epfino, for example, have some facilities for data entry restrictions and cross-checking. They are excellent for the static purpose for which they are designed, but rather crude in comparison with those provided by a dedicated database tool. Second, it requires considerable experience to achieve maximum benefit. Third, it does not help at all in the major undertaking of data exploration. The authors have analysed a large scale survey using a statistical package and a database in collaboration. Their experience is reported here. | lld:pubmed |
