Physical intervention in schools guidance: consultation analysis

Annex A: Methodology

For the purposes of this report, the terms “response” and “respondent” were defined as explained below. This explains the difference between the person or persons who provided the response, and the content of the response provided.

• Response - The individual comments received from a respondent for the consultation;
• Respondent - The group, individual, organisation or sector that submitted the response to the consultation.

As part of the consultation analysis, respondents were categorised into the relevant group, individual, organisation or sector to which they relate. The respondent categories are as follows:

• Individual
• Local authority
• Education stakeholder
• Third sector
• Other professional stakeholder
• Other organisation

Discussion on approach for quantitative data

The consultation included quantitative questions where the respondents were asked to select their answer from a “Yes”, “No” choice framework. The numbers of responses in each category were counted and are presented in section 3.2. Where respondents did not provide an answer for a question, a “Not answered” category is also analysed.

Discussion on approach for qualitative data

A coding framework was developed and used to analyse the qualitative data received for the consultation, which allowed for a consistent approach to be taken when analysing the responses.

A coding framework is a way of categorising the text to establish and identify key themes and patterns in qualitative comments. This involves reviewing the comments received, searching, and identifying concepts and finding relations between them.

Creating the coding framework involved reviewing the questions asked in the consultation to identify potential broad themes and issues that were likely to be raised in the responses. These potential broad themes were then tested using a small sample of the responses received. The coding framework was designed to reflect the level of detail that would be helpful in considering any amendments to the draft guidance and to allow for a useful grouping of responses. Once the broad themes were established, new sub themes were added, as required, in order to capture an additional level of detail. The coding framework was updated as the analysis progressed. Responses were not restricted to one broad theme or sub theme per response. There was no limit on the number of sub themes that could be assigned per response. The consultation responses included generic comments or observations that did not directly relate to the questions set out. The coding framework provided for this. Below is a sample from the coding framework of a broad theme and sub themes used to undertake the consultation analysis:

• Additional clarity needed in specific sections
  • Length
  • Structure
  • Prevention
  • Staff-led withdrawal
  • Seclusion
  • Training

Once all responses were coded, they were subject to further review to identify key themes. Where a theme is referenced as a key theme in this report, it was identified as a theme raised 18 times or more in response to individual consultation questions.

Limitations of the analysis

It is also important to note the limitations of this approach and the assumptions present throughout the analytical process. The voice of the respondent, the analyst and the reader interpreting the coding framework are all present. The coding framework evolved based on themes emerging from the responses and was not restricted to pre-set categories identified by the analyst. The analysis of the responses is limited by the context of the questions and the draft guidance. Care should be taken when using a coding framework that attributes numerical values to qualitative data. The framework used allows key themes to emerge, however these are not quantifiable in the same way as data gathered from quantitative questions. As a result, the themes are not presented in a way which attributes a number of responses to them. It is accepted that this will not provide detail on the volume of responses within each key theme. Instead, key themes are explored to reflect where a range of comments touch on the same issues. Further, the analysis is based on interpretation of a limited number of responses from a relatively small group of respondents and care should be taken in interpretation of these themes at a population level.