ARCHIVE DOCUMENTATION STUDY NO: 4036 STUDY TITLE: National Diet and Nutrition Survey: People Ages 65 Years and over 1. VALIDATION AND CONVERSION: 1.1 CONTENT VALIDATION: Level of validation The file ndnsqre received a 100 per cent check on its data. The rest of the files received the normal 40 variable plus 10 per cent. The following is clarification about the meaning of variables MAIN22F1 - MAIN22F5 (Supplement product licence number) and MED1LIC - MED16LIC (Medical product licence number) These above variables refer to the product licence number. Most prescribed and non-prescribed supplements and medicines have a product licence number which uniquely identifies the manufacturer and the product. The first four digits refers to the manufacturer, the second four digits identify the product. If the interviewer was not able to see the packaging for medicines or supplements this information was not recorded. Missing value labels MEMQ (Do you remember what year it is now?) This variable contains 92 cases or '3', the depositor was unable to provided an answer for its meaning or why it was such a large number. NSQH9B1-NSQH9B4 (In which countries have you lived?) The National Centre only used codes grouped below. European Community = 001-012, 900, 901 Non European Community = 014-096, 902 Africa = 204-395, 910-913 America = 400-529, 920-922 Asia = 600-743, 930-932 Australia, Oceania and other countries = 800-890, 940 Any country, N.E.C=990 Not answered = 999 Missing documentation Page 37 of the main questionnaire is a duplicate of page 36, consequently question 76 is missing from the questionnaire. The depositor was requested on numerous occasions to provide this page, but has yet to do so. We also requested that the depositor sent us a copy of the show cards as it was felt that this would have aided in the analysis of the data. The depositor has yet to provide this. 1.2 CONVERSION PROCESS (for data and documentation) inputs/outputs, A number of dat and csv files were deposited these files were converted to SPSS A large a mount of paper documentation was received. This was scanned in and converted to PDF. Word files and an ascii text file were also converted to PDF. tools used, SPSS Acrobat Capture 1.3 SUMMARY OF VALUE ADDED PRODUCTS: Data Set-up files were written for the following raw files: ndnsqre diary day plate food nutrient Wherever possible value labels and question text was added, full variable lists, descriptives and some frequencies run. Documentation The questionnaires and section for were annotated to indicate which variables related to which question. PDF files were created 2. CONSTRAINTS ON THE RELIABILITY AND/OR USABILITY OF THE DATASET 2.1 FILENAME CHANGES Filenames for user documentation and data are changed by the Data Archive for file management purposes. It is possible that original filenames have been referred to in the body of the documents. We provide below a table showing any filename changes that have occurred. Documentation |------------------------------------------------------------------------| | Filename | Description | PDF Filename | |--------------|------------------------------------------|--------------| | arch_c1.doc } Section 1 - Background } | |--------------}------------------------------------------} | | arch_c2.doc } Section 2 - Methodology } | |--------------}------------------------------------------} a4036uab.pdf | | arch_c3.doc } Section 3 - Questionnaire & Diary Coding } | |--------------}------------------------------------------} | | arch_c4.doc } Section 4 - Database Structure & D Vars } | |--------------|------------------------------------------|--------------| | appenf.txt } Derived Variable Specification } a4036ucb.pdf | |--------------|------------------------------------------|--------------| 3. NOTES, USEFUL TIPS AND HINTS More detailed information on the value labels, (unless provided in the coding notes) can be found by running a DISPLAY DOCUMENT command on the data files provided with this study. An example program is provided below: Title 'Example of Display Document & Dictionary File handle imp/name='/path/example.por' Import file=imp Display Documents Display Dictionary Finish DATA The depositor sent a larger number of data files in a number of formats, these formats include dat, csv with sps set-up files, # delimited files and a POR file. Users will need to be sent either: 1) Both raw and derived *.dat and corresponding set-up files with all the *.csv and their corresponding set-up files AND the # delimited files. OR 2) Both raw and derived *.por files. SECOND EDITION INFORMATION - NOVEMBER 2001 ------------------------------------------ Addition of data from Oral Health Survey (dental data) ------------------------------------------------------ For the second edition of this dataset (November 2001), data from the four waves of the Oral Health part of the survey was deposited in ascii format. SPSS portable files have been created from these original files, one for each of the 39 rows of data (as described in the User Guide), plus a survey weights file (dentwt.por) and a dietary variables file (dietsums.por), specific to the dental data. The 'row' files (row1.por to row39.por) contain data for each row from all four waves of the survey - a wave identifier variable (WAVE) has been added to the files to identify waves 1, 2, 3 or 4. Variable names for the 'row' files have been taken where possible from the row details in the User Guide - otherwise they reflect the variable label and are self-explanatory. Variable labels have been added to the files, but not value labels, which can be obtained easily from the questionnaire and examination record in the User Guide (Part 5 - Oral Health Survey). Users are strongly advised to order the .por format files, as the original ascii files are arranged by respondent and row, and so may prove difficult to use. Some anomalies have been found in the data and documentation for the Oral Health Survey: 1. In files dentwt.por and dietsum.por, the serial numbers (subject code) are 5-digit unique numbers. In all other .por files, the serial number may also have an alphabetical character at the end of the unique number. 2. Isolated wild or out-of-range codes were found for many questionnaire variables throughout the 'row' data files. 3. Some waves were found to contain an extra line of data for each row, whereas others have one row less than the depositor states: Row 3 (file row3.por) - Wave 1 contains 228 cases, not 229. Row 6 (file row6.por) - Wave 2 contains 222 cases, not 221. Row 7 (file row7.por) - Wave 2 contains 220 cases, not 221. Row 16 (file row16.por) - Wave 2 contains 222 cases, not 221. Row 17 (file row17.por) - Wave 2 contains 220 cases, not 221. 4. Where open text responses have been recorded, such as variables ATTLST3 'What trouble' and ATTRX2 'Type of treatment' in row 5 (file row5.por), numeric codes seem to have been allotted to the responses. There is no information given about these codes in the User Guide. 5. Some errors in the data row specifications in the Oral Health Survey User Guide have been noted: a. Row 1 (page 73 of Appendix 8) - there is an extra variable in the data between SAT3 and DENU1 - this is the screening question 5, which has been given the variable name SAT5. - the last variable in Row 1 has been incorrectly labelled 'Dentures - Drop Speak' - it is in fact screening question 11, which has been given the variable name SCREEN11 instead of DEN17. 'Dentures - drop speak' appears in Row 2 (variable DEN18). b. Row 5 (page 79 of Appendix 8) The unlabelled variable at Column 11 is Screening question 42, and has been given the variable label SCREEN42. c. Rows 34 and 38 (pages 108 and 112 of Appendix 8) Columns 30-32 should read 'Upper left' 6,7 and 8, not 'Upper right' on both pages.