Diversity in political representation in Scotland: data improvement project proposal

4. Designing the Questionnaire

4.1 Approach

In designing the questionnaire for this exercise, we have sought to take into account a range of factors as set out below:

• The questions should allow comparison of the profile of election candidates with the national picture.
• This exercise may be able to make a contribution to the development of a standardised questionnaire and approach to reporting that can be used by others including political parties. Seeking feedback now is intended to ensure the questionnaire is as useful as possible, and future-proof as far as possible for future data collection exercises.
• Response rates for similar surveys have been low. In order to seek to maximise the likelihood of candidates responding, the questionnaire will be as short and straightforward to complete as possible.
• Candidates may find some questions more sensitive than others. Instructions will be provided to respondents making clear that questions are voluntary and if there are any questions they do not want to answer, they can just go on to the next one.
• Questions will be designed to be proportionate and gather only as much detail as required to obtain a picture across the protected characteristics.
• Accessibility - ensuring the questionnaire is available in formats that meet the needs of respondents.

The proposed questionnaire has been developed along the lines of equality monitoring forms and diversity monitoring forms that are commonly used in a range of settings, with the addition of potential questions on parliamentary experience, socio-economic status/background and caring responsibilities. While diversity monitoring forms are commonly used in a range of settings, there is not a single existing suite of agreed questions to use for this purpose^[46], and indeed one of the weaknesses noted in the EHRC's report about the currently available data on election candidates, was the lack of consistency in data collection, and the associated "fragmented picture across protected characteristics with many gaps"^[47].

We have reviewed key sources of guidance on collecting diversity data, explored the approaches taken by different surveys, and sought advice from a number of equality and diversity stakeholders to date. To enable comparison with the national picture and consistency with similar international surveys we specifically looked at questions in the following surveys:

• 2022 Census^[48];
• Scottish Household Survey (SHS)^[49];
• Improvement Service Survey of Councillors^[50];
• Welsh Local Government Candidates Survey^[51];
• Comparative Candidates Survey^[52].

4.2 Proposed questions

In order to collect data on the diversity of election candidates and elected representatives in Scotland the questionnaire proposes to ask candidates and elected members questions on the following protected characteristics:

• age;
• sex;
• sexual orientation;
• gender reassignment;
• disability;
• race;
• religion or belief.

To capture additional information on the diversity of candidates and elected representatives, the questionnaire also proposes to ask questions on:

• socio-economic status;
• previous parliamentary experience;
• caring responsibilities.

Annex A sets out the proposed questions and response options in full as well as the rationale for each question choice. Where possible, the questions proposed in the questionnaire reflect either or both of the 2022 Census and the annual Scottish Household Survey.

To ensure that the form and subsequent data is as useful as possible for partners and stakeholders, we would welcome comments and feedback on what is proposed.

4.3 Making the questionnaire available in different formats

Engagement with stakeholders to date has indicated that an approach that provides both an electronic and a paper-based option for responding will best meet candidate's needs. The questionnaire will therefore be provided alongside an information leaflet as follows:

1. As an online survey, accessible via a QR code and hyperlink provided on the paper and printable questionnaire taking candidates to an electronic version of the survey to enable them to complete it on their phones or other electronic devices;

2. As a paper form, leaflet and return envelope given out with nomination papers by ROs (marked as non-mandatory); and

3. As a printable form and leaflet available to download online and print (marked as non-mandatory) for those not travelling to collect nomination papers from their council office.

We will continue to consider further options to ensure the form and process for responding are as accessible as possible, and would welcome comments and feedback on this.

4.4 Facilitating analysis of both candidates and elected representatives in one exercise – collecting names

The exercise aims to enable analysis and reporting both on the diversity of all candidates for election in May 2022, as well the diversity of the cohort of councillors elected. To avoid the need for two separate data collection exercises, and the associated risk that newly (re)elected councillors may not be keen to fill in the same form a second time in the aftermath of the election, we are proposing to ask for candidate names as part of the exercise.

Under the proposed approach candidates would be asked to provide their name with their online or paper questionnaire response. Prepaid sealable envelopes would be provided to candidates with the paper forms, intended to provide both a straightforward and free means of returning their questionnaire, but also some reassurance that their data is being returned privately in envelopes that will not be opened until the data entry stage.

Once analysts receive the sealed envelopes and electronic survey data they would attach a unique identifier to each candidate name and to their diversity data. The names would then be separated from the candidate's responses and stored separately in restricted electronic storage. Paper questionnaire responses would be stored in secure locked storage and destroyed as soon as quality assurance of the data entry is complete. As a result, only the unique identifiers would be linked to candidates' diversity data.

Using unique identifiers is standard practice with surveys of this nature and enables trained analysts / researchers to keep track of which responses belongs to which candidate without using the candidates' names.

Following the election, the successful candidates will be marked on the list of candidate names and unique identifiers. The record of successful and unsuccessful candidates by unique identifier only would then be linked to the diversity data so that the diversity dataset can then be analysed by either all candidates or by successful and unsuccessful candidates. Candidate names and linking information would at all times be stored securely and separately from diversity data, and only a small number of analysts undertaking the data entry would have access to candidates' names and response data.