Database Preparation Services

Database Statistical Analysis

LTI offers two standard statistical reports (RecStat and DupStat) to help the library better understand the characteristics of its database. Custom reports can be prepared on request.

RecStat reports, which are generated before and after deduping, are valuable in estimating present and future database and index mass storage requirements. The first part of a RecStat report is a frequency count matrix listing bibliographic format (books, serials, AV, etc.) versus record transaction status (Produce, Update, Cancel, etc.). This report provides a quick overview of the library's database by listing total record counts and percentages falling in each category.

RecStat also presents record counts by four-character holding library codes, including the number of records and holdings fields in which multiple holding library records are found. Other RecStat statistics summarize the average size of a record, along with the number of characters in the largest and smallest record. A bar chart displays the number of records falling into different size ranges. The same format is used to summarize the number of fields per record. A final field use summary is arranged by tag number. It lists each field and the number of times it occurs, the average number of times it occurs per record; the number and percentage of records in which the field occurs one or more times; and the minimum, maximum, and average number of characters in the field.

LTI's DupStat report offers detailed statistics of the number and percentage of duplicate records in a library's database. It lists the total number of records, the number and percentage of unique and duplicate records; the frequency of duplicate record groups; and the processing status of each record within groups of 2, 3, or 4 duplicate records. The DupStat report is used to identify duplicate record patterns and to determine which groups of duplicate records should be reviewed manually prior to choosing a deduping method. As a check on the effectiveness of the deduping process, DupStat reports are produced before and after deduping.