Guidance

DWP research report guidance

Updated 12 August 2022

© Crown copyright 2022

This publication is licensed under the terms of the Open Government Licence v3.0 except where otherwise stated. To view this licence, visit nationalarchives.gov.uk/doc/open-government-licence/version/3 or write to the Information Policy Team, The National Archives, Kew, London TW9 4DU, or email: psi@nationalarchives.gov.uk.

Where we have identified any third party copyright information you will need to obtain permission from the copyright holders concerned.

This publication is available at https://www.gov.uk/government/publications/dwp-research-reports-style-guide/dwp-research-reports-guidance

Published: 1 June 2009.

This guidance applies to all Department for Work and Pensions (DWP) research reports which will be published in one of the following collections:

If you need to use a different approach to the one set out in this guidance, contact your DWP project manager to discuss alternative approaches before submitting the first draft of your report. Alternative presentation of GSR reports is possible (often published as part of the ad hoc research collection). However, the principles of this style guide should be followed where possible.

1. Structure

All research reports follow the same structure.

Use the research report template to write your research report, which should include the following sections:

  1. Title page
  2. Publication information
  3. Code of Practice for Statistics compliance statement (this only applies to research which relies heavily on statistics)
  4. Executive summary
  5. Contents
  6. Acknowledgements
  7. Author details
  8. Glossary
  9. Abbreviations
  10. Summary
  11. Main report
  12. Appendices
  13. References

1.1 Title page

Include:

  • the research report title you have agreed with your DWP project manager – the title should be:
    • 65 characters or less, including spaces (aim for ≤65 characters and use a subtitle if you need to include extra information)
    • meaningful when shown in a list of search results
    • written in plain English
  • the DWP and Government Social Research (GSR) logos – if the research involves other government departments or agencies, contact your DWP project manager to arrange for other logos to be added
  • the research report publication month and year
  • author(s) and research organisation(s)

1.2 Publication information

Include:

  • the DWP research report series number or DWP ad hoc research report number – your DWP project manager will provide this
  • author(s)
  • copyright information
  • the ISBN – your DWP project manager will provide this

If the research was externally commissioned, you must also include the following disclaimer:

‘Views expressed in this report are not necessarily those of the Department for Work and Pensions or any other government department.’

1.3 Code of Practice for Statistics compliance statement

If your research relies heavily on statistics, consider including a voluntary statement explaining how the research complies with the 3 pillars of the UK Statistics Authority Code of Practice for Statistics. If your research doesn’t rely heavily on statistics, you may still consider how these pillars have been addressed. The pillars are:

  • trustworthiness
  • quality
  • value

Trustworthiness

If the research has been externally commissioned, explain that contractors work with DWP to understand the aims of the research. However, the design, delivery and analysis:

If analysis has been carried out internally, explain how this is made transparent, for example by including research instruments and data tables.

Consider explaining that although research findings are shared with ministers and other officials before publication, this is done to promote the value of the research to the department. Ministers have no editorial role.

Quality

Explain:

  • the analytical approach
  • how the analysis and report have been quality assured

Value

Explain DWP’s response to the research, for example:

  • how the department has acted or will act on the research findings
  • how processes have been improved or will be improved as a result of the research

1.4 Executive summary

The executive summary should be no more than 500 words. It should help people decide whether they want to read the report, and should summarise:

  • contents
  • findings
  • recommendations

1.5 Contents

Include a table of contents and a list of any figures and tables.

Use numerals (for example 1, 2, 3) for page numbers.

1.6 Acknowledgements

You should:

  • if applicable, state that the research was commissioned by DWP
  • acknowledge the role of DWP staff in the research

1.7 Author details

List authors in alphabetical order.

Use no more than 50 words per author when providing author information.

1.8 Glossary

Write the report in plain English. Where you cannot avoid the use of technical or scientific terms, include these in the glossary.

1.9 Abbreviations

Include a list of abbreviations used in the research report.

1.10 Summary

This is the section which ministers, senior policy advisors and other stakeholders are most likely to read. It should be no more than 3,000 words, and people should be able to understand the summary independently of the main report.

Include:

  • the aims of the research
  • a short description of the research methodology
  • main findings and conclusions – include references, for example “The majority of claimants thought staff were knowledgeable (section 3.2)”
  • any policy recommendations

1.11 Main report

Include:

  • a description of the research and its aims
  • background information
  • a guide to the research methodology
  • information about any special conventions used in presenting findings, for example the system for referring to a series of surveys
  • analyses and findings

When producing both the summary and main report, consider the likely consumer of the publication and when, where and why they will use the product. The report should be designed for the consumer, not the creator.

It is important not to assume knowledge.

1.12 Appendices

Include:

  • detailed technical information, for example sample frame (there may be some cases circumstances where a technical annex should be published as a separate document, alongside the main report)
  • detailed methodological information
  • sampling method
  • response rates
  • questionnaires, topic guides and other field documents – if a document is particularly long, it does not need to be included in the research report, but should be made available on request

Your DWP project manager might ask you to provide copies of quantitative research documentation and cleaned data so that they can be archived.

1.13 References

1.13.1 Bibliography

List references alphabetically by author in a single bibliography.

Below are examples of the preferred bibliography reference format for the most common types of documents that you may wish to reference.

Examples

Journal Article: Author, A.N. and Writer, A. (year of publication) ‘Title of paper: subtitle of paper’, Title of Journal, volume number, issue number, pages

  • Stern, P.C. (2000) ‘Towards a coherent theory of environmentally significant behaviour’, Journal of Social Issues, 56, 3, 407-424

Electronic Journal Article: Author, A.N. and Writer, A. (year of publication) ‘Title of paper: subtitle of paper’, Title of Journal, volume number, issue number, pages. Retrieved from Sage journals: sociological rersearch online

Government Research Reports: Author, A.N. and Writer, A. (year of publication) ‘Title of paper: subtitle of paper’, organisation/department name. Retrieved from: URL

Book: Author, A.N. (year of publication) Title of book: subtitle of book, Town [and State if U.S.]: Publisher

  • Ritchie, J., Lewis, J., Nichols, C.M. and Ormston, R. (2014) Qualitative research in practice, Thousand Oaks [CA]: SAGE Publications

Electronic Book: Author, A.N. (year of publication) Title of book: subtitle of book, Town [and State if U.S.]: Publisher. Retrieved from: URL

Chapter/Paper in an Edited Book: Author, A.N. (year of publication) ‘Title of paper: subtitle of paper’, in Writer, A. (ed.), Title of book (pages), Town [and State if U.S.]: Publisher

  • Pennycook, G. and Thompson, V.A. (2017) ‘Base-rate neglect’, in Pohl, R.F. (ed.), Cognitive illusions: intriguing phenomena in thinking, judgment and memory (pages 44-61), London: Routledge

1.13.2 Within Text and Footnote Citations

Within text and footnote citations should contain enough information for the reader to identify the reference in the bibliography. For references within the text or footnotes, the following style is preferred (in full first time and truncated afterwards):

One Author: (Author/Organisation, year)

Two Authors: (Author and Author, year)

Three Authors: (Author and others, year)

Read more about references.

2. Style and content

2.1 Style

Read Writing for GOV.UK before you write your research report. It includes guidance on:

You should also read the Government Digital Service (GDS) style guide. It is an A to Z of GOV.UK style, and includes guidance on:

Your research report drafts and final report should not require style changes, grammar or spelling corrections upon return to the DWP project manager.

2.2 Headings

In the word document, set headings correctly using ‘styles’ (i.e. don’t just bold and increase font size) and ensure that they are nested correctly. For example, Heading 3 should only be used if it’s preceded by Heading 2 and Heading 1 before that.

2.3 Font and Spacing

Research reports should use single spacing, Arial and size 12. These are highly accessible and so will help anyone using a PDF, regardless of whether they have specific access needs.

2.4 Quotations

Use quotations to illustrate a point, not to make a point. Do not use quotations instead of analysis.

Use double quotes (“”) for direct quotations.

Include quotations of one sentence or less in normal text. Indent longer quotations.

2.4.1 Sources

Documentary sources

For quotations from documentary sources, include the source in brackets immediately after the quotation.

Attribute quotations from published sources in the footnotes.

Verbatim quotations

Explain the system for attributing verbatim quotations from interviews before the first quotation.

Include the source on the line following the quotation. Explain where the source fits into the sampling scheme, for example by stating age, sex and benefit status.

For example:

“I first claimed Jobseeker’s Allowance last year. That was after I lost my job at the factory.”
(Woman, 20 to 39, current recipient)

2.4.2 Selecting quotations

Consider the overall balance of quotations in your research report and the weight it gives to the evidence. Make sure the balance of positive and negative quotations is an accurate reflection of the evidence. For example, if a report is overwhelmingly positive about an initiative, more positive than negative views should be shown in quotations.

When you select quotations, take into consideration that the most memorable quotations:

  • are often the ones that represent extreme views, and they can distort the research evidence
  • often represent a particular position, for example a positive view of a service, that can create a false picture of the evidence

You should therefore consider:

  • including less memorable but more representative quotations
  • explaining that a quotation represents an extreme opinion, for example “such views could be very strongly held, with one respondent going as far as to say…”
  • including more than one quote to give an indication of the range of views, where different views are expressed on an issue

Links should be embedded within the text, as it allows the reader to click directly to the referenced material. For example, “in the report Thriving at Work: a review of mental health and employers, it was found that…”.

As in the above example, use description language for embedded links. Do not use generic links in sentences , such as “you can view the report here”.

2.6 Footnotes

Considering the user need for footnotes is important; footnote use is a matter of judgement for project managers, but it is advised to limit the use of footnotes where possible. If the footnote just refers to a link, this should be embedded in the body of the text, as described in the ‘Links’ section. If you can include a reference in the body of the text (e.g. as with Harvard and A.P.A referencing, as opposed to Oxford referencing), this is advised.

If adding a footnote, insert after the punctuation mark. For example, “Of those men in couples with children who did not have a job, only 22% declared that they would not want a job if one were available.”[footnote 1] This does not apply when it is obvious that the footnote reference number only refers to the word it is next to.

Within text and footnote citations should contain enough information for the reader to identify the reference in the bibliography and should be formatted as described in the ‘References’ section.

2.7 Numbers, fractions and percentages

You should:

  • include percentages after fractions where they’re written out in the text, for example “three-quarters of Income Support claimants (74%) were dissatisfied”
  • round figures of 0.5 or above to the nearest whole number (except in special cases, for example where showing slowly changing indices)
  • round large numbers where appropriate
  • use words rather than figures to make a point where readers are likely to misinterpret the figures (percentages over 100% are often misinterpreted)
  • avoid comparisons of different-sized fractions, for example thirds and fifths

More guidance on numbers, fractions and percentages is available in the GDS style guide.

2.8 Tables

When you use a table:

  • number tables using the chapter number, followed by a full stop and the number of the table within that chapter, for example, the second table to appear in Chapter 3 will be ‘Table 3.2’
  • summarise the information and the main findings in the text to ensure accessibility
  • use concise and descriptive titles, which tell the key message of the table in one sentence. A sub-heading should then provide more detail on the information displayed
  • indicate the cell contents, for example ‘row percentages’, ‘column percentages’, ‘multiple response’ or ‘average weekly income’
  • include the unit of measurement, for example ‘£’, in the column or row heading, not in the cells of the table
  • use right align
  • use commas for large numbers
  • start with 0 for values less than 1 (e.g. 0.87, not .87)
  • consider effective rounding
  • ensure data is ordered appropriately
  • give only the unweighted bases in the table where the difference between weighted and unweighted numbers is small. Where the weights are large, and the 2 sets of numbers are very different, include both the weighted and unweighted bases
  • describe the base underneath the table, for example ‘Base: all respondents (1,520)’, where possible
  • quote the source beneath the table, if applicable
  • make sure tables are consistent throughout the report. For example, use a consistent level of precision

Use the following conventions in tables, and explain them before the first table, for example:

Symbol Meaning
* Less than 0.5 per cent
0 No observations
- Category not applicable
.. Data not available
[ ] Numbers in square brackets are frequencies based on less than 50 observations

Do not:

  • use lines with cells
  • use percentages where the base falls below 50, either report the actual frequency or indicate that the base was too small for a figure to be calculated
  • use letters in table headings, for example ‘Table 3.2 a’
  • use large, complicated tables – instead consider breaking up the data and including it in smaller tables. Particularly avoid merged cells where possible, as these are difficult to format in HTML
  • refer to tables as appearing ‘above’, ‘below’ or on a certain page number, refer to them by number only, for example ‘see Table 3.2’

2.9 Images and charts

If you include images or charts in your research report, you should:

  • number diagrams or figures using the chapter number, followed by a full stop and the consecutive number of the diagram within that chapter, for example the second diagram to appear in Chapter 3 will be ‘Figure 3.2’

  • explain the main information conveyed by the image or chart in the report text. There is no need to include alternative text (sometimes called ‘alt text’) if the information it conveys is explained in the report text, or if the image is decorative (e.g. a logo). If alternative text is necessary (i.e. it is not feasible to describe the content of the image in the body text), then briefly describe the content of the image to make sense as though there’s no image on the page; for example, write the alternative text as “the health minister tweeted…”, instead of “an image of a tweet by the health minister”. For information on how To add alternative text in Word, see the DWP research report template.
  • use concise and descriptive titles which tell the key message of the chart in one sentence. A sub-heading should then provide more detail on the information displayed
  • include a key for graphs and diagrams, where appropriate
  • label axes fully (with units of measurement, if appropriate)
  • ensure axis are labelled horizontally for ease of the reader
  • consider employing the use of chart annotations and labelling in colour to boost audience understanding and reduce cognitive load
  • use gridlines sparingly
  • ensure data is ordered appropriately
  • include the sample base, unweighted, where appropriate
  • quote the source beneath the diagram, if applicable
  • size images at 960x640 pixels for HTML formatting purposes.
  • insert images or charts in the report, or submit them separately as an Excel spreadsheet
  • ensure the style of figures and diagrams is consistent throughout the research report for publication

Do not use 3D charts or pie charts unless absolutely necessary – they are difficult to read and interpret.

2.10 Colour

Colour must not be relied upon as the only means of conveying a message to the reader; reports should remain as effective if viewed in black and white. To ensure the effective use of colour, consider referring to the ONS colour guidance on presenting information with regards to colour blind audiences using shades and tone. To reduce pressure to the eyes, shades should be used in logical order.

All in-house research and analysis should use DWP colours. These can be found in the AC SharePoint (‘AC Brand - chart and table templates’)

2.11 Case studies

If using case studies for data presentation purposes (e.g. to illustrate a particular feature of an intervention or outcome), consider presenting as follows to better align with HTML formatting:

  • if it’s just a single paragraph, the case study can be presented in a grey box between paragraphs of the main text.
  • if longer than this, it can be presented in plain text format, with headings to indicate that it’s a case study.

2.12 Response rates

Many of the samples for DWP’s research come from benefit records. This section explains how you should present response rates when using these records.

Two figures are of interest to DWP:

  • the response rate for those in scope of the study (in the sampling frame), to give an indication of the risk of bias in the achieved sample
  • the response rate for those in scope of fieldwork (in the contacted sample), to give an indication of field performance

Always calculate response rates from the number of in-scope cases in the sample. You can calculate this by:

(number of achieved interviews x 100) ÷ (number of cases sampled – number of ineligible cases)

Ineligible cases are those which are found not to be eligible for the study. For example:

  • people sampled for a survey of families who were found not to have children
  • people not getting Income Support in a survey of Income Support recipients

When you have found the number of cases which were in scope of the study, calculate the response rate for those in scope of fieldwork by:

(number of achieved interviews x 100) ÷ (number of cases in scope of study – (number of invalid cases + number of cases issued to interviewers))

Cases may be invalid for the following reasons:

  • respondent moved from given address, and not traceable
  • respondent moved from given address, and traceable to an area outside fieldwork
  • respondent not known by occupants of given address
  • respondent died
  • address derelict or empty
  • address not traced or not complete
  • address business or industrial (except in special cases, for example surveys of Child Support Agency clients, who often give their work address for contact, or surveys of employers)

Valid cases for benefit record-based samples, which you should use to calculate the response rate, fall into the following categories:

  • successful interviews
  • doorstep refusals
  • no contact after agreed number of call-backs
  • appointments broken after agreed number of call-backs
  • interview stopped by respondent
  • respondent away during fieldwork
  • respondent incapable of doing interview, for example due to illness of incapacitation

Do not include cases which are not used at all, because the required number of interviews is achieved before they are issued, in the fieldwork response rate.

Population (N) Population in scope of study (%) Population in scope of fieldwork (%)
Number sampled 2,222 N/A N/A
Ineligible 180    
       
In scope of study 2,042 100 N/A
       
Cases not issued to interviewers (opt-outs) 222    
       
Invalid cases:      
Respondent moved out of fieldwork area 10    
Respondent moved, untraceable 280    
Respondent not known by occupants 30    
Respondent died 10    
Address derelict or empty 60    
Address not traced/not complete 10    
Address business/industrial 10    
Total 410    
       
In scope of fieldwork 1,410 69 100
       
Non-contact after agreed number of call-backs:      
With anyone at the address 80    
With named person 110    
Total 190    
       
Refusals:      
All information refused 10    
Personal refusal 100    
Broken appointment, no re-contact 40    
Proxy refusal 30    
Total 180    
       
Other reasons for non-interview:      
Respondent away during fieldwork period 30    
Respondent incapable of interview (for example due to health problems) 20    
Inadequate English or refused interview in English 10    
Interview terminated by respondent 10    
Total 70    
       
Response rate 970 48 69
Population (N) Population in scope of study (%) Population in scope of fieldwork (%)
Number sampled 2,222 N/A N/A
Ineligible 180    
       
In scope of study 2,042 100 N/A
       
Cases not issued to interviewers (opt-outs) 222    
       
Invalid cases:      
Business moved out of fieldwork area 10    
Business moved, untraceable 280    
Business not known by occupants 30    
Out of business or ceased trading 10    
Address derelict or empty 60    
Address not traced or not complete 10    
Residential address 10    
Total 410    
       
In scope of fieldwork 1,410 69 100
       
Non-contact after agreed number of call-backs:      
With anyone at the address 80    
With appropriate job-holder 110    
Total 190    
       
Refusals:      
All information refused (for example switchboard refusal) 10    
Personal refusal 100    
Broken appointment, no re-contact 40    
Proxy refusal 30    
Total 180    
       
Other reasons for non-interview:      
Respondent away during fieldwork period 40    
Inadequate English or refused interview in English 10    
Interview terminated by respondent 20    
Total 70    
       
Response rate 970 48 69

3. Standalone summary

You should provide a standalone summary of your research report if it is going to be published in the DWP research report collection. You do not need to provide a standalone summary if it is an ad hoc research report.

Use the DWP standalone summary template to write your summary. It should be no more than 3,000 words but may be significantly less than this limit (e.g. 2,000 or less). Contact your DWP project manager to agree a date to submit it.

See an example of an HTML standalone summary.

4. Submitting your report

Before you submit your report to your DWP project manager:

  • proofread all document(s) and ensure that comments and track changes have been removed
  • remove company and organisational logos
  • ensure the main report is available in one Word file
  • ensure graphs and figures are available in a format agreed with your DWP project manager – in the case of graphs and figures this will normally be the spreadsheet from which the charts were originally generated
  • ensure the main report is available in one Word file
  • ensure your report and, where applicable, standalone summary meets all accessibility requirements by following the HTML guidance that your DWP project manager will provide you with

  1. In contrast, over 50% of women in the same situation said they would not want a job if one were available. ↩

Back to top