NATIONAL SURVEY OF LOCAL GOVERNMENT CANDIDATES, 2003
3. RESPONSE RATES AND REPRESENTATIVENESS
A total of 2013 responses were received from the 2976 unsuccessful candidates, representing an overall response rate of 67.6%. For a postal survey of this nature this is a highly satisfactory figure and compares favourably to the response rate achieved in the survey of councillors (60.5%). However, as table 1 shows, the overall response rate hides considerable variation across councils - ranging from a low of 53.6% in North Lanarkshire to a high of 88.9% in Eilean Siar (although the number of candidates in the latter council was small). Generally there was a slightly higher response from more rural authorities, but it is reassuring that the response rate did not fall below 50% in any of the 32 councils.
Table 1: Response rates by council
Council | Candidates | Responses | % Responses |
Aberdeen City | 149 | 100 | 67.1 |
Aberdeenshire | 145 | 116 | 80.0 |
Angus | 88 | 65 | 73.9 |
Argyll and Bute | 73 | 55 | 75.3 |
Clackmannanshire | 31 | 21 | 67.7 |
Dumfries and Galloway | 94 | 66 | 70.2 |
Dundee City | 101 | 56 | 55.4 |
East Ayrshire | 76 | 51 | 67.1 |
East Dunbartonshire | 74 | 57 | 77.0 |
East Lothian | 77 | 62 | 80.5 |
East Renfrewshire | 58 | 46 | 79.3 |
Edinburgh, City of | 223 | 145 | 65.0 |
Eilean Siar | 18 | 16 | 88.9 |
Falkirk | 65 | 46 | 70.8 |
Fife | 212 | 146 | 68.9 |
Glasgow City | 263 | 150 | 57.0 |
Highland | 91 | 56 | 61.5 |
Inverclyde | 58 | 34 | 58.6 |
Midlothian | 60 | 36 | 60.0 |
Moray | 52 | 37 | 71.2 |
North Ayrshire | 73 | 48 | 65.8 |
North Lanarkshire | 125 | 67 | 53.6 |
Orkney Islands | 16 | 11 | 68.8 |
Perth and Kinross | 104 | 83 | 79.8 |
Renfrewshire | 113 | 74 | 65.5 |
Scottish Borders | 65 | 43 | 66.2 |
Shetland Islands | 16 | 12 | 75.0 |
South Ayrshire | 73 | 42 | 57.5 |
South Lanarkshire | 171 | 109 | 63.7 |
Stirling | 71 | 51 | 71.8 |
West Dunbartonshire | 46 | 31 | 67.4 |
West Lothian | 95 | 69 | 72.6 |
TOTAL | 2976 | 2013* | 67.6 |
* Note: This figure contains 12 candidates whose council is unknown.
In terms of representiveness by political affiliation, as table 2 shows the survey received a positive response from all parties (and from independents). The response rate from Liberal Democrat candidates, at 81.2%, was particularly high, but it is encouraging that over 60% of candidates from all four major political parties completed the questionnaire (the response rate from SSP candidates was only slightly below 60%).
Table 2: Response rates by party
Party | Responses | % Responses |
Independent | 184 | 71.0 |
Conservative | 432 | 63.8 |
Labour | 297 | 72.3 |
Liberal Democrat | 406 | 81.2 |
SNP | 498 | 63.4 |
SSP | 182 | 58.3 |
Other 2 | 14 | 51.9 |
TOTAL | 2013 | 67.6 |
Finally, how representative is the sample in terms of gender? As the gender profile of all candidates is known (collated from lists of persons nominated), a comparison between responding candidates and all candidates is possible. As table 3 shows, the gender profile of responding candidates almost exactly matches that of all candidates - 30.1% of the sample were female compared to 30.0% of all candidates. With this overall high response rate, it's perhaps not surprising that the gender profile of respondents so closely mirrors that of all candidates.
Table 3: Response rate by gender
Gender | All Respondents | % | All Unsuccessful Candidates | % |
Male | 1408 | 69.9 | 2083 | 70.0 |
Female | 605 | 30.1 | 893 | 30.0 |
Given the high response rates across councils and parties, and the fact that the gender profile of respondents almost exactly matches that of all unsuccessful candidates, there is little to be gained from re-weighting the sample to take account of differential response rates. Moreover, many questions will be analysed by political affiliation and clearly there is no intrinsic bias due to the differential response rates across parties in such tables.