讲解 Advanced Qualitative Research, 2023/24 SOCS0016调试SPSS

Advanced Qualitative Research, 2023/24 SOCS0016

Assessment Guide

AQM SOCS0016

There is one summative essay of 3,000 words due on 22.04 before 1pm.

Assessment description:

For this assessment, you are expected to choose one method of data analysis that has been covered in this module (grounded theory; discourse analysis, narrative analysis or thematic analysis) and use this approach to analyse one data set.

In addition to the 3000 words, you should include appendices with evidence of how you analysed the data, such as a table of codes and themes, with illustrative data text (for thematic analysis) or examples of small stories and how they were analysed to generate your findings (eg for narrative analysis). These are not included in the word count.

You are required to use ‘open’ or ‘safeguarded’ data drawn from the UK Data Service: https://ukdataservice.ac.uk/

There are excellent guidance videos on how to register and find appropriate qualitative data on the UK Data Service website: https://ukdataservice.ac.uk/find-data/

We recommend that you choose from one of the four following data sets. In each case, please sample and analyse a minimum of five transcripts from within each dataset:

1. Scott, S. and McDonnell, L. (2020) Asexual lives: Everyday experiences, relationships and stories of becoming: https://reshare.ukdataservice.ac.uk/851821/
Data type: 45 Interview transcripts

2. Davies, K. (2021) Brexit and Everyday Family Relationships, 2019-2020 https://reshare.ukdataservice.ac.uk/854710/
Data type: 10 Interview transcripts

3. Brown, P. (2017) Losing and finding a home: A life course approach: https://reshare.ukdataservice.ac.uk/850721/
Data type: Approx 18 interviews with service professionals, approx. 80 interviews with people who have lived experience of homelessness

4. Laffer, Alexander (2023). Attitudes Towards Emotional Artificial Intelligence Use: Transcripts of Citizen Workshops Collected Using an Innovative Narrative Approach, 2021: https://reshare.ukdataservice.ac.uk/855688/
Data type: 10 transcripts of focus groups

However, you are also welcome to explore the UK Data Service holdings and find other appropriate qualitative data sets to analyse. Please check with one of the module leads (Alison or Katherine) to verify if the data set you have identified is appropriate.

For the essay, you must:

1. Construct a research question that is relevant to your chosen data set

2. Choose one of the four data analysis methods covered in this module and use it to analyse the chosen dataset

3. Write up your analysis explaining your research question, analytic approach and findings. To do this, you will need to include a short literature review of the subject area to help support your research question but we do not expect this to be comprehensive: the main focus is demonstrating your ability to analyse data to answer a clear research question

Assessment rationale:

The aim of this assessment is to give you an opportunity to demonstrate your understanding and practical skill of a specific research analysis method. By actively using the analytic method you will develop a more rigorous familiarity with it beyond just describing it. This assessment also gives you the chance to work with real research data and experience the ‘messiness’ of analysis first hand: part of the skills we want you to demonstrate is to be able to draw out clear, coherent conclusions which speaks to the complexity of the data.

How to plan your time:

To excel, we strongly recommend you start browsing the suggested data sets (or others) before Reading Week, and construct a research question by Week 5. Reading data sets takes time and requires you to take notes as you read, so make sure to factor in at least a week’s worth of work to complete this step. This process will be discussed in Week 3 on Secondary Analysis. In the seminars for Week 5, we will give some formative feedback on your research questions so you will be able to check if you are going in the right direction.

You will also need to do a bit of additional reading to provide a short literature review to support your research question: this section of your report does not need to be huge but should reflect a understanding of crucial features of the project (e.g., if you use the dataset on asexuality you need to demonstrate a basic grasp of what asexuality is and what gaps in the research into asexual experience is). If nothing else, a solid literature review will help you to develop your analysis by pointing you towards areas of interest, tensions between theory and empirical experience, or provide useful concepts to test with your data. Reading Week would be a good time to conduct this additional reading and pull together the literature review.

The next task is to then explore the different methods of data analysis: we will introduce these one by one in the lectures and seminars for Weeks 6-10. If you know you have a particular interest in a specific method already, then we encourage you to do additional reading for the method even prior to the week we cover it in class: this will give you time to think about the method in conjunction with the data you have. If you are not sure which approach you want to apply yet then the first step is to explore the key readings for each session and take good notes in classes as we cover each approach: spend time each week trying each method out experimentally on your chosen data set, so that you have a sense of how they work and what they illuminate in your data.

We suggest that you start drafting your report no later than Week 9, in order to give yourself at least a week to conduct the analysis of the data and a week to write up your findings.

How to lay out your report:

Your report requires a short introduction which should introduce the structure, data set and method of analysis that will be used throughout, and indicate key findings. You may also want to include your research questions in the introduction or include them in the body of your essay along with the literature review.

The literature review and background information should follow the introduction, and be used to motivate your research questions. This section is where you can frame. your approach in a particular way (if relevant), such as taking a feminist, post-structural or post-colonial approach. This is also where you can situate your approach to secondary analysis and how you intend to make your analysis trustworthy. In other words, draw on the sessions in the first part of the course to help inform. this section.
To be successful, your research questions must be relevant, answerable and qualitative: by relevance, think about what can be reasonably learnt from your dataset, in conjunction with your literature review on the area; by answerable, make sure that the scope of your research questions are relatively limited and focussed, and can be addressed by the data contained within the dataset you have (so, don’t have a question about the political process of Brexit if your dataset is about Brexit and family relationships: while the topic is broadly aligned, the data won’t help you answer a question looking at political process); and by qualitative, ensure that your questions are exploring experience, feelings, values or meaning. Qualitative questions are usually framed along the lines of “What is the experience of…” and “How is meaning is given to…”   compared to quantitative questions which tend to be framed along the lines of, “To what extent does…” and “What causes…?” Try to limit the number of research questions you generate to make your analysis and discussion of findings clearer.

The bulk of the word count should be given over to the data analysis and findings. You can label these sections however seems most appropriate. Within them, it is important to describe the method of analysis you are using, drawing on the methodological literature to support your understanding; and then to explain clearly how you applied the method to the dataset. How you present this will depend on the method you choose and your own critical thinking around what your reader needs to know in order to appreciate the analysis. The second part is to then demonstrate the findings or results of your analysis. For example, for thematic analysis, this would mean presenting and explaining the themes you have generated; for discourse analysis, it will mean introducing and justifying the discourses you have identified, and so on. You will need to draw on quotations from the source material to help support your analysis, and contextualise how you reached those findings by connecting your discussion back to the literature review, and by anchoring your findings back to your research question to show how your analysis answers your question. This is a crucial step for the analysis to succeed. This means you may end up tweaking your research question a bit to align with the findings of your analysis as you write up; this is not unusual in research but make sure your research questions are still relevant, answerable and qualitative.

The last part of the body of the report is the conclusion. As part of this section, give some assessment of how successful you feel your analysis was, and on what basis, and account for any challenges you met in the process (for example, are there sections of data that couldn’t fit into your schema, or was there an outlier in your data who disrupted the analysis, etc.).  It is important to demonstrate critical analytical skills as applied to the work you have conducted. This does not mean denigrating your assessment, but rather a careful consideration of the limitations and your learning in the process of undertaking the analysis, both of the topic and the analytical method.

The final section of the report is not included in your word count, as it may be very long. This section is the Appendix (or appendices if you include more than one). We would like you to include evidence of your analysis process in the appendices: again, this will look different for each method used. For thematic analysis, it may be the codebook generated (if codebook thematic analysis was used), or an example of a coded passage of text. For discourse analysis, it may look like pages of highlighted text showing how you identified the construction of a discourse. For narrative analysis, it may look like highlighted sections of text showing how you located the structure of narratives; and so on.

In summary, your structure should look a bit like this:

1. Introduction

2. Short literature review

3. Data analysis

4. Findings

5. Conclusion

6. Appendices (excluded from the word count)

Things to beware of:

In analysis, there are a range of issues to be aware of regardless of the specific approach you chose. We will discuss these in the process of the module but it is worth highlighting to you here that we will be looking to see ways in which you have avoided: over-simplification; reductive analysis; overclaiming; and summarising without analysis.

Over-simplification

All analysis is simplification to some extent, as you are finding ways to communicate complex reality in a condensed, clear way, and you are drawing out specific patterns of features of the data that your analysis has created. However, the way you write about your findings is important, as is remaining accountable to the complexity of the original source data. It can be tempting to exclude or ignore complicated data points, or ‘outlier’ respondents who don’t seem to fit the mould that the rest of your participants neatly fit into. To ‘tidy up’ your data in this way through exclusion, or to write as if all participants did fit the mould is to risk oversimplifying the complexity in such a way that it no longer speaks to the whole of the phenomenon discussed. Do not be afraid to point out where your analysis does not fit the data, and hold yourself accountable to what is actually happening: even if it means telling a less comprehensive-sounding story.

Reductive analysis

As above, all analysis is in some sense reducing complexity in order to tell a story from the data that is comprehensible. However, reductive analyses are those which point to one or two (themes/discourses/narratives/theories) and claim that the whole of the phenomenon is captured by these alone, sweeping away the possibility of other intervening issues, or denying the explanatory power of alternative approaches. The result is to present a naïve analysis which is blind to the complexity of the source data: when writing up your analysis, do not be afraid of positioning your findings as partial, or contributing to wider understanding rather than giving a total and wholly comprehensive discussion.

Overclaiming

Related to both of the above is the problem of over-claiming from the analysis. Qualitative research tends to sit within constructivist or interpretive paradigms which are content with the idea that research itself is one view, or one construction among many. When writing, therefore, we tend to use terms like “could be, may be” (rather than “is”), “suggests, implies, suggests” (rather than “shows or is”), and “some, partial, limited, in this case, within this data” (rather than “all, always, in all cases”). This use of tentative language reduces the risk of overclaiming from the data.

Summarising without analysis

This is the hardest thing to avoid. Whether you are looking for narratives or themes, discourses or a new concept, it is perfectly possible to read through data and draw out interesting points and summarise them without actually analysing them. Over the course of the module we will push you to think about this in more detailed ways, but in brief here: to summarise, you only need to describe what is said in the data accurately. This in itself can be very difficult and time consuming, and to do it well is a research skill in itself. To analyse, you need to look at your summaries and account for the why, or to draw it back into conversation with your research question in a way that brings new understanding out of the data explored.

Questions?

On the moodle page there is an assessment Q&A forum where you can post any questions: this can be really helpful as it builds a resource all students can then look back to and see the answers.

Please email either Katherine or Alison with specific questions about datasets and so on.

Marking Rubric: