Writing qualitative research papers

writing qualitative research articles papers

We’ve actually talked about communicating qualitative research and data to the public before, but never covered writing journal articles based on qualitative research. This can often seem daunting, as the prospect of converting dense, information rich studies into a fairly brief and tightly structured paper takes a lot of work and refinement. However, we’ve got some tips below that should help demystify the process, and let you break it down into manageable steps.

 

Choose your journal

The first thing to do is often what left till last: choose the journal you want to submit your article to. Since each journal will have different style guidelines, types of research they publish and acceptable lengths, you should actually have a list of a few journals you want to publish with BEFORE you start writing.

 

To make this choice, there are a few classic pointers. First, make sure your journal will publish qualitative research. Many are not interested in qualitative methodologies, see debates about the BMJ recently to see how contested this continues to be. It’s a good idea to choose a journal that has other articles you have referenced, or are on a similar topic. This is a good sign that the editors (and reviewers) are interested in, and understand this area.

 

Secondly, there are some practical considerations. For those looking for tenure or to one day be part of schemes that assess the quality of academic institutions by their published work such as the REF (in the UK) or PBRF (in New Zealand) you should consider ‘high impact’ or ‘high tier’ journals. These are considered to be the most popular journals in certain areas, but will also be the most competitive to get into.

 

Before you start writing, you should also read the guidance for authors from the journal, which will give you information about length, required sections, how they want the summary and keywords formatted, and the type of referencing. Many are based on the APA style guidelines, so it is a good idea to get familiar with these.

 


Describing your methodology, literature review, theoretical underpinnings

When I am reviewing qualitative articles, the best ones describe why the research is important, and how it fits in with the existing literature. They then make it clear how the researcher(s) chose their methods, who they spoke to and why they were chosen. It’s then clear throughout the paper which insights came from respondent data, and when claims are made how common they were across respondents.

 

To make sure you do this, make sure you have a separate section to detail your methods, recruitment aims and detail the people you spoke to – not just how many, but what their background was, how they were chosen, as well as eventually noting any gaps and what impact that could have on your conclusion. Just because this is a qualitative paper doesn’t mean you don’t have to say the number of people you spoke to, but there is no shame in that number being as low as one for a case study or autoethnography!

 

Secondly, you must situate your paper in the existing literature. Read what has come before, critique it, and make it clear how your article contributes to the debate. This is the thing that editors are looking for most – make the significance of your research and paper clear, and why other people will want to read it.

 

Finally, it’s very important in qualitative research papers to clearly state your theoretical background and assumptions. So you need to reference literature that describes your approach to understanding the world, and be specific about the interpretation you have taken. Just saying ‘grounded theory’ for example is not enough – there are a dozen different conceptualisations of this one approach.
 

 

Reflexivity

It’s not something that all journals ask for, but if you are adopting many qualitative epistemologies, you are usually taking a stance on positivism, impartibility, and the impact of the researcher on the collection and interpretation of the data. This sometimes leads to the need for the person(s) who conducted the research to describe themselves and their backgrounds to the reader, so they can understand the world view, experience and privilege that might influence how the data was interpreted. There is a lot more on reflexivity in this blog post.


How to use quotations

Including quotations and extracts from your qualitative data is a great feature, and a common way to make sure that you back up your description of the data with quotes that support your findings. However, it’s important not to make the text too dense with quotations. Try and keep to just a few per section, and integrate them into your prose as much as possible rather than starting every one with ‘participant x said’. I also like to try and show divergence in the respondents, so have a couple of quotes that show alternative view points.

 

On a practical note, make sure any quotations are formatted according to the journal’s specifications. However, if they don’t have specific guidelines, try and make them clear by always giving them their own indented paragraph (if more than a sentence) and clearly label them with a participant identifier, or significant anonymised characteristic (for example School Administrator or Business Leader). Don’t be afraid to shorten the quotation to keep it relevant to the point you are trying to make, while keeping it an accurate reflection of the participant’s contribution. Use ellipsis (…) to show where you have removed a section, and insert square brackets to clarify what the respondent is talking about if they refer to ‘it’ or ‘they’, for example [the school] or [Angela Merkel].

 


Don’t forget visualisations

If you are using qualitative analysis software, make sure you don’t just use it as a quotation finder. The software will also help you do visualisations and sub-set analysis, and these can be useful and enlightening to include in the paper. I see a lot of people use an image of their coding structure from Quirkos, as this quickly shows the relative importance of each code in the size of the bubble, as well as the relationships between quotes. Visual outputs like this can get across messages quickly, and really help to break up text heavy qualitative papers!

 


Describe your software process!

No, it’s not enough to just say ‘We used Nvivo’. There are a huge number of ways you could have used qualitative analysis software, and you need to be more specific about what you used the software for, how you did the analysis (for example framework / emergent) and how you got outputs from the software. If you did coding with other people, how did this work? Did you sit together and code at one time? Did you each code different sources or go over the same ones? Did you do some form of inter-rater reliability, even if it was not a quantitative assessment? Finally, make sure you include your software in the references – see the APA guides for how to format this. For Quirkos this would look something like:

 

Quirkos Software (2017). Quirkos version 1.4.1 [Computer software]. Edinburgh: Quirkos Limited.

 

Quirkos - qualitative analysis software

 


Be persistent!

Journal publication is a slow process. Unless you get a ‘desk rejection’, where the editor immediately decides that the article is not the right fit for the journal, hearing back from the reviewers could take months or even a year. Ask colleagues and look at the journal information to get an idea of how long the review process takes for each journal. Finally, when you get some feedback it might be negative (a rejection) or unhelpful (when the reviewers don’t give constructive feedback). This can be frustrating, especially when it is not clear how the article can be made better. However, there are excellent journals such as The Qualitative Report that take a collaborative rather than combatitative approach to reviewing articles. This can be really helpful for new authors.

 

Remember that a majority of articles are rejected at any paper, and some top-tier journals have acceptance rates of 10% or less. Don’t be disheartened; try and read the comments, keep on a cycle of quickly improving your paper based on the feedback you can get, and either send it back to the journal or find a more appropriate home for it.

 

Good luck, and don’t forget to try out Quirkos for your qualitative analysis. Our software is easy to use, and makes it really easy to get quotes into Word or other software for writing up your research. Learn more about the features, and download a free, no-obligation trial.

 

 

Does software lead to the homogenisation of qualitative research?

printing press homogenisation qualitative method

 

In the last couple of weeks there has been a really interesting discussion on the Qualrs-L UGA e-mail discussion group about the use of software in qualitative analysis. Part of this was the question of whether qualitative software leads to the ‘homoginisation’ of qualitative research and analysis. As I understand it, this is the notion that the qualitative sphere is contracting from diverse beginnings, narrowing to a series of commonly used and accepted methods of collection and interpretation. For example, the most popular are probably semi-structured interview transcripts coupled with some type of framework based interpretation. Are more and more researchers using qualitative research churning out work using the same research? Is modern qualitative technology leading to a unified outputs like the introduction of the printing press, or helping increasing the accessibility of the discipline?


While I do see some evidence of trends emerging in the literature and research articles, I do not see them as inevitable, or feel that alternative approaches have been relegated, or that software need be a force for homogenisation.


Actually, I see a lot of similarities in this debate with a keynote talk on conformity in qualitative research by Professor Maggie MacLure at the ICQI conference last year. Referencing Deleuze, Nietzsche and the Greek Myths, she described the need to balance the dichotomy of two of the sons of Zeus in Greek legend: Apollo and Dionysus. Dionysus represents, chaos, emotion (and excess drinking of wine) while Apollo masters truth, rational thinking and prophecy. One can argue that following Apollo can lead to homogenisation, while too much Dionysus in your research can lead to chaos and a difficulty in drawing meaningful conclusions (especially with the wine drinking, although many researchers I know would disagree on this important point when writing up research).


However, a little creativity is important, especially at the point of choosing your methodology. In qualitative research, you can use arts-based research, using participant creation of drawings, games or even pottery as data. There are real challenges in keeping the richness of these creative methods alive through the analysis process: how do you analyse a drawing by a participant? Yet it’s rarely enough to just look at transcripts of respondents talking about their creations, and ignoring the art work itself. So take a pinch of the creative to ward against homogenisation: the excellent overview on Creative Research Methods by Helen Kara is a great place to start.


But what about the analysis and qualitative software? Can this be creative and unique as well?


I would argue that it can – especially with certain tools. I think there is a tendency for software to ‘lead’ users into particular behaviours and approaches, which is why users should look at the Five Level QDA approach advocated by Woolf and Silver and decide how they want to analyse their data before choosing a software package. But most software is very flexible. Even tools like Atlas.ti that was originally designed for grounded theory can be used for other theoretical approaches (Friese 2014). However you can still see this legacy in the design, for example the difficulty in creating a hierarchical coding structure in Atlas.ti remains today.


The design methodology for Quirkos was to create a very simple qualitative software tool that allowed people to use it in anyway they wanted. And in my experience from 3 years in running a qualitative software company, I can assure you that there is little risk of homogenisation in software users! Users occasionally share their projects with me to get advice on a problem, and I can see people using the features in ways we never envisaged! I also get lots of emails in my Inbox with suggestions on how we can make small tweaks to allow people to use Quirkos in different ways. The demand from the users is not to adopting the same approach over and over again, but being able to customise the software to their own needs and ways of working. And again, I can assure you these approaches are more diverse than I ever imagined.


And what about the argument that software creates mechanical and thought-less analysis? Well, I think this is a risk, and I’ve written about the discipline that users need to avoid this. But I think that any reductive analytical process risks becoming automatic, and thus removing the richness of the qualitative data. Even a pen and highlighters approach to analysis can become automatic and brainless if not done with care, and when re-reading data the eye can skip to the brightly coloured sections, sometimes missing vital context.


Ironically there is also some homogenisation in the software industry itself. Many scholars including Fielding and Lee (1998) have talked about ‘Creeping featurism’ and a trend of software packages to become more similar and (complex) as they add tools and functionality from each other. They tend to have similar interfaces, and function in ways that often seem very similar to the new user. Now, a fan of any one qualitative software package will quickly let you know how superior X is to Y because of a subtle aspect of the layout, and how easy it is to work in a particular way. Again this seems to evidence that software itself does not lead to homogenisation of approaches.


There are more than a dozen qualitative software packages actively developed at the moment, and between them they offer a fantastic variety of conceptual and practical approaches to data coding and management. For most people I speak to, the choice of software is bewildering, just like the variety of methods that can be used in qualitative research. I hope that new students are led so that, rather than being shoehorned into a particular approach, they are excited by the dizzying heights of possibility in qualitative research.


If you would like to give the unique Quirkos experience a try, we have a free trial you can download so you can see if the simple, visual and colourful approach is right for your qualitative research. And as ever, if you have any questions, feel free to get in touch with us at support@quirkos.com.