Published  |  Last updated

Quick Answer: A mixed methods research paper is judged on how plainly it reports six things: why both data types were needed, the design, each method, where and how the strands were integrated, any limit one method places on the other, and what mixing added.1 Give the qualitative analysis, the quantitative analysis and the mixed methods analysis their own sections, then show the integration in a joint display, table or graph.2 The MMAT, a tool built for appraising studies in systematic mixed studies reviews, asks whether the integration was interpreted and any divergence explained.3

How to Write a Mixed Methods Research Paper: What the Standards Ask For

Most advice on mixed methods is about design. This page is about the paper itself: what to write, in which section, and how to make the integration of the two strands visible. For choosing a design and planning integration before you collect data, see our guide to combining qualitative and quantitative approaches.

The reason to report integration explicitly is documented. A 2008 review of mixed methods studies funded by the Department of Health in England identified 118 studies and obtained proposals or final reports for 75. Researchers mainly ignored the mixed methods design and described only the separate components, and judgements about integration could rarely be made because no attempt at integration was reported.4 That review formed the basis of GRAMMS.1

The Six GRAMMS Reporting Items

These are the six GRAMMS items as listed in the 2026 protocol to update them.1 The third column is where we suggest each one goes in a paper.

Swipe sideways to see every column.

#GRAMMS itemWhere it usually goes
1Describe the justification for using a mixed methods approach to the research question.End of the Introduction, repeated in one sentence in Methods: say why one data type alone could not answer the question.
2Describe the design in terms of purpose, priority and sequence of methods.Methods, first paragraph: name the design, say which strand leads, and give the order.
3Describe each method in terms of sampling, data collection and analysis.Methods, one subsection per strand.
4Describe where integration has occurred, how it has occurred and who has participated in it.Mixed methods analysis subsection and the integrated Results.
5Describe any limitation of one method associated with the presence of the other method.Methods or Discussion, only where one strand actually affected the other.
6Describe any insights gained from mixing or integrating methods.Discussion: what the two strands together showed that neither showed alone.

GRAMMS is more than 15 years old. The 2026 protocol describes a methodological review and a modified Delphi process to produce GRAMMS 2.0 and says the integration item needs more detail.1 A PubMed search on 31 October 2026 for a published GRAMMS 2.0 guideline returned only that protocol, so GRAMMS stays the working checklist.

What APA JARS-Mixed Adds, Section by Section

APA’s mixed methods standards (JARS-Mixed) were developed in 2018 and are used together with the JARS-Qual and JARS-Quant standards.2,5 They are written by manuscript section, which makes them the better guide to the order of your paper.2

Swipe sideways to see every column.

SectionWhat JARS-Mixed asks for
TitleReference the mixed methods, qualitative methods and quantitative methods used, and avoid words that signal one strand only, such as “explore” or “understand” (qualitative) and “determinants” or “correlates” (quantitative).
AbstractSpecify the type of mixed methods design and indicate the participants or data sources, the analytic strategy, the main results and the major implications.
IntroductionMay convey barriers in the literature that suggest a need for both qualitative and quantitative data.
Methods: designGive a rationale for collecting both kinds of data and the added value of integrating the results from the two.
Methods: data sourcesIdentify the qualitative and quantitative sources, their characteristics and the relationship between them, in the order of the procedures. A table of data type, timing and participants helps. Describe data as open-ended or closed-ended rather than words versus numbers.
Methods: recruitment and analysisDiscuss recruitment for each strand separately. Devote separate sections to the qualitative analysis, the quantitative analysis and the mixed methods analysis.
ResultsIndicate how the results were mixed (discussion, joint display tables, graphs, or data transformation such as turning themes into counts) and mirror the design sequence.
DiscussionMirror the same sequence and reflect on the implications of the integrated findings.

The design decides the order of your write-up. The MMAT manual describes the three core designs, and JARS-Mixed says how each is integrated.3,2

Swipe sideways to see every column.

DesignWhat it meansHow the strands are integrated, and the write-up order
ConvergentQualitative and quantitative components usually run at the same time to examine the same phenomenon.Merged. If both were gathered together, either strand may be reported first.
Explanatory sequentialQuantitative results from phase 1 inform the qualitative phase 2, which explains them.Connected. Report quantitative first, then qualitative.
Exploratory sequentialQualitative findings from phase 1 inform the quantitative phase 2, which tests or generalises them.Connected. Report qualitative first, then quantitative.

Where to Put the Integration: Narrative, Data Transformation and Joint Display

Fetters, Curry and Creswell describe integration at three levels: study design, methods, and interpretation and reporting. At the reporting level it happens through narrative, data transformation and joint display.6 These are the options for reporting integration: the first three are narrative approaches, and the last two are other reporting-level approaches.

Swipe sideways to see every column.

ApproachWhat it meansOur suggestion for when to use it
Weaving (narrative)Qualitative and quantitative findings are written together, theme by theme or concept by concept.One theme or concept links the strands and the reader needs numbers and words side by side.
Contiguous (narrative)Both sets of findings sit in one report, but in different sections.The strands answer different sub-questions, or the journal format keeps them apart.
Staged (narrative)Results of each step of a multistage study are reported as they are analysed, often in separate papers.Each stage is its own paper. State the link to the companion papers.
Data transformationCoded qualitative themes are converted into counts, then integrated with the quantitative data.You want to put qualitative codes into the same table or graph as survey results.
Joint displayRelated data are organised in a figure, table, matrix or graph.You want the reader to see the meta-inference in one view.

The same article defines the “fit” of integration as how well the findings cohere. Confirmation means both sources agree, expansion means they diverge but add complementary aspects, and discordance means they are inconsistent.6 In a review of 19 joint displays in three journals, the most common types were statistics-by-themes and side-by-side comparisons.7

What the MMAT Asks of Your Integration

The Mixed Methods Appraisal Tool is built for appraising studies in systematic mixed studies reviews, so it shows what an appraiser will look for in your paper. Its mixed methods category has five criteria, quoted here as worded in the 2018 manual.3 Our guide to quality assessment in mixed-methods systematic reviews covers the review side.

  1. 15.1 Is there an adequate rationale for using a mixed methods design to address the research question?
  2. 25.2 Are the different components of the study effectively integrated to answer the research question? The manual gives joint displays as an example of bringing the data together, and asks when integration occurred.
  3. 35.3 Are the outputs of the integration of qualitative and quantitative components adequately interpreted? The manual links this to meta-inference, which shows the added value of a mixed methods study over two separate studies.
  4. 45.4 Are divergences and inconsistencies between quantitative and qualitative results adequately addressed? The manual says it is not sufficient to only report divergences; they need to be explained.
  5. 55.5 Do the different components of the study adhere to the quality criteria of each tradition of the methods involved? The manual says the overall quality of a mixed methods study cannot exceed the quality of its weakest component.

The manual discourages calculating one overall score from these ratings and advises presenting each criterion’s rating instead.3

A Real Joint Display, Redrawn in R

To see these ideas in a finished paper, we used an open-access study of caregivers of children with cancer who had a fever with very low white cells (febrile neutropenia, FN). Smeallie and colleagues ran a convergent mixed methods study at one centre. Their survey had 13 quantitative items and 8 open-ended items, completed by 28 patient and caregiver dyads, and they also abstracted 27 FN episodes from medical charts.8 The paper is CC BY 4.0, which allowed us to redraw its numbers. Their joint display tables were first printed incomplete and were republished in a corrigendum that restored the qualitative results and meta-inferences.9

The integrated findings

The article reports three themes and a meta-inference for each. The table below rebuilds the joint display in the statistics-by-themes style from numbers and quotes printed in the article and corrigendum.8,9 The fit column is our reading using Fetters’ terms; the authors do not label fit.

Swipe sideways to see every column.

ThemeQuantitative strand (as reported)Qualitative strand (quote)Meta-inference (article) and fit (our reading)
Family burdenGroup means: household tasks 3.0 for early discharge vs 1.9 for standard admission; 2.7 low-risk vs 2.1 high-risk; time with partner 2.8 vs 2.1.“[The FN hospital admission] kept our family apart for 3 extra days. My aunt had to watch my boys as my husband had to work.” (CG 04; low-risk FN, standard admission)FN episodes negatively affect the whole family unit. Fit: confirmation, as both strands show more strain with longer or higher-risk episodes.
Treatment location94% of coded statements about home care positive (15/16); 78% of those about hospital care negative (25/32); satisfaction with care 4.6 early discharge vs 4.1 standard admission.“We got [a]round the clock care so her blood count numbers usually are better maintained within the hospital then at home.” (CG 14; high-risk FN, standard admission)Caregivers weigh benefits and risks of each location and see home treatment as most beneficial when the outpatient team supports it. Fit: confirmation overall, with one reported voice that points the other way.
Medical management87% of coded statements about disease management negative (20/23).“Mood and appetite are always worse in the hospital.” (CG 26; high-risk FN, early discharge)Caregivers want a comprehensive approach to managing the child and family. Fit: expansion, as the open-ended items add symptoms and tailoring that the closed items do not measure.

The pairing of survey means with themes is ours: the article’s Table 2 pairs the family burden theme with a bar graph of response counts per survey item, which we could not reproduce from the text, so we show the group means printed in the text instead. Quotes are reproduced from the article under CC BY 4.0.

Horizontal stacked bars of coded caregiver statements by sentiment: home care 15 of 16 positive (94%), hospital care 25 of 32 negative (78%), disease management 20 of 23 negative (87%)
The qualitative strand turned into counts (data transformation), redrawn from the article’s corrigendum. Reported: 87%, 78% and 94%. The other counts are totals minus the reported count, computed in R, assuming each coded statement was negative or positive. Data from Smeallie et al. (CC BY 4.0).
Dumbbell chart of mean survey scores: early discharge vs standard admission 3.0 vs 2.6 overall, household tasks 3.0 vs 1.9, child's independence 3.1 vs 2.3, satisfaction with care 4.6 vs 4.1; low-risk vs high-risk 2.8 vs 2.7, 2.7 vs 2.1, 2.8 vs 2.1
The quantitative strand: group means printed in the article text, with differences computed in R. The article prints no significance tests for these comparisons. Data from Smeallie et al. (CC BY 4.0).

Where the paper reports its integration

This is the GRAMMS question in practice. The figure maps the four places a reader can find the integration step in the article. The mapping to GRAMMS items is our reading of the article text.

Four-step flow of where one open-access paper reports integration: Methods 2.1 design and planning, Methods 2.6 mixed methods analysis, Results 3.3 mixed method results, Discussion 4.1 and 4.2 main findings, with GRAMMS items 2, 4 and 6
Where Smeallie et al. report integration, by section number, with the GRAMMS items each touches (Editverse reading). Section text from Smeallie et al. (CC BY 4.0).

Swipe sideways to see every column.

GRAMMS itemHow the case paper reports it (Editverse reading of the article text)
1 JustificationThe Introduction names a gap, few in-depth descriptions of dyads’ experiences especially in North America, and states the aim. Section 2.3 says the 8 qualitative items were added to thematically expand upon the quantitative items.
2 DesignSection 2.1 names an observational, convergent mixed methods study. We did not find a statement of priority or sequence.
3 Each methodSurvey with 13 quantitative and 8 qualitative items, recruited to thematic saturation (Sections 2.3 and 2.4), and chart abstraction of 27 episodes (Section 2.5).
4 IntegrationWhere and how: a joint display of integrated data collection at planning (2.1) and a joint display analysis that draws meta-inferences (2.6). We did not find who carried out the integration stated in the Methods.
5 One method’s limit due to the otherSection 4.2 lists sample size, single-site recruitment and unproctored joint responses. None is framed as an effect of one method on the other.
6 Insights from mixingSection 4.2 says the design allowed the relationships between positive and negative factors to be illustrated, and Figure 1 of the article shows them.

Computed by Editverse in R

  • Group differences in the survey means: +0.4, +1.1, +0.8 and +0.5 (early discharge minus standard admission) and +0.1, +0.6 and +0.7 (low-risk minus high-risk).
  • Chart counts re-derived from the reported numbers: 14 of 27 episodes low-risk is 51.85% (the article prints 51.8%), 13 of 27 high-risk is 48.15%, 11 of 27 early discharges is 40.74%, 9 of 11 early discharges low-risk is 81.82% and 5 of 16 standard admissions low-risk is 31.25%.
  • Reconciling counts: the article says 31 dyads were approached with an 87% response rate and reports 28 dyads. 87% of 31 is 27.0 and the chart analysis has 27 episodes, so a reader has to work out how the three numbers relate.

What this case shows: a paper that reports integration in four places, gives each theme its own meta-inference, and includes a voice that does not fit the majority. What it does not show: that home care causes better outcomes. The sample is 28 dyads from one centre, and the article prints no significance tests for the group comparisons. The practical lesson is to proof the joint display with the same care as the abstract: in this paper the three tables lost their qualitative results and meta-inferences in production.9

A Section-by-Section Checklist Before You Submit

  1. 1Title and abstract. Reference both strands in the title, avoid words that signal only one, and give the design type in the abstract.
  2. 2Introduction. End with the reason both data types are needed and what integration will add.
  3. 3Methods. Name the design, give the order of procedures in a table, and describe each strand separately.
  4. 4Analysis. Write three subsections: qualitative, quantitative and mixed methods analysis, the last saying how the strands were merged or connected.
  5. 5Results. Report each strand, then the integrated findings in a joint display with a meta-inference per theme and any divergence explained.
  6. 6Discussion. Mirror the same order and state what mixing showed that neither strand showed alone.

Need the Analysis Run or Checked? Consult Editverse

Editverse subject-matter experts work to COPE, ICMJE and GPP 2022 guidance, with institutional and country-specific ethics followed.

Editverse logo

Editverse

Manuscript Writing

Both strands collected and the paper still unwritten?

You send: your data from both strands, your design notes and your target journal.

You get: a manuscript written and edited by PhD experts, with statistical review, journal selection and submission support.

When: Your project is finished and the paper is not.

Manuscript Writing Services · $1,500

Write my paper →

Editverse logo

Editverse

Statistical Analysis

Need the quantitative strand run and reported?

You send: your dataset and your research questions.

You get: the quantitative analysis run in SPSS, R or Stata and reported, ready to sit beside your themes in a joint display.

When: You need numbers you can place next to your qualitative findings.

SPSS/R/STATA Statistical Analysis · $400

Analyse my data →

Editverse logo

Editverse

Journal Selection

Not sure which journals take mixed methods papers in your field?

You send: your abstract, field and any indexing your institution requires.

You get: a ranked list of 5-10 journals with scope fit, indexing coverage (PubMed, Scopus, Web of Science), review timelines and open access fee information, and the rationale for each.

When: You need a journal that fits a two-strand paper.

Scientific Journal Selection Services · $115

Get my journal shortlist →

Free Research Tools from Editverse

Frequently Asked Questions

What is the structure of a mixed methods research paper?

It follows the usual sections, with the two strands and their integration made visible in each. The APA JARS-Mixed standards ask for a rationale for collecting both data types, separate analysis sections for the qualitative strand, the quantitative strand and the mixed methods analysis, results sections that mirror the design sequence and show how the strands were integrated, and a discussion that reflects on the integrated findings.

What are the GRAMMS items?

GRAMMS (Good Reporting of A Mixed Methods Study) lists six things to describe: the justification for a mixed methods approach, the design in terms of purpose, priority and sequence, the sampling, data collection and analysis of each method, where integration occurred, how it occurred and who took part in it, any limitation of one method associated with the other, and any insights gained from mixing the methods.

What is a joint display?

A joint display organises related quantitative and qualitative data in one figure, table, matrix or graph so the two can be read together. In a review of 19 joint displays in three journals, the most common types were statistics-by-themes and side-by-side comparisons.

Do I need a separate mixed methods analysis section?

JARS-Mixed says to devote separate sections to the qualitative analysis, the quantitative analysis and the mixed methods analysis. The mixed methods analysis describes how the results were merged in a convergent design or connected in an explanatory or exploratory sequential design.

What is a meta-inference?

It is the overall interpretation drawn from integrating the qualitative and quantitative findings. The MMAT asks whether these outputs of integration are adequately interpreted, because they show the added value of a mixed methods study over two separate studies.

What should I do when the qualitative and quantitative results disagree?

Explain the disagreement, not only report it. The MMAT says divergences must be addressed and names reconciliation, initiation, bracketing and exclusion as strategies. Fetters and colleagues call an inconsistency between the two data sources discordance.

Should the quantitative or the qualitative results come first?

Follow the design. JARS-Mixed says the findings should mirror the design sequence: quantitative first in an explanatory sequential design, qualitative first in an exploratory sequential design, and either first when both were gathered at the same time.

Is there a published update to GRAMMS?

A protocol for GRAMMS 2.0, a methodological review followed by a modified Delphi process, was published in JMIR Research Protocols in March 2026. A PubMed search on 31 October 2026 for a published GRAMMS 2.0 guideline returned only that protocol, so the six-item GRAMMS list remains the one to use.

References

  1. Munce S, Fàbregues S, Hong QN, et al. Updating the Good Reporting of a Mixed Methods Study (GRAMMS) reporting guidelines: protocol for a methodological review and modified Delphi process. JMIR Res Protoc. 2026;15:e82364. doi:10.2196/82364. Open access, CC BY 4.0. pubmed.ncbi.nlm.nih.gov/41841629/
  2. American Psychological Association. Mixed Methods Design Reporting Standards (JARS-Mixed), Table 1. APA Style (page last updated January 2024). apastyle.apa.org/jars/mixed-methods
  3. Hong QN, Pluye P, Fàbregues S, et al. Mixed Methods Appraisal Tool (MMAT), version 2018: user guide (last update 1 August 2018). Registration of Copyright #1148552, Canadian Intellectual Property Office. mixedmethodsappraisaltoolpublic.pbworks.com/f/MMAT_2018_criteria-manual_2018-08-01_ENG.pdf
  4. O’Cathain A, Murphy E, Nicholl J. The quality of mixed methods studies in health services research. J Health Serv Res Policy. 2008;13(2):92-98. doi:10.1258/jhsrp.2007.007074. pubmed.ncbi.nlm.nih.gov/18416914/
  5. Levitt HM, Bamberg M, Creswell JW, et al. Journal article reporting standards for qualitative primary, qualitative meta-analytic, and mixed methods research in psychology: the APA Publications and Communications Board task force report. Am Psychol. 2018;73(1):26-46. doi:10.1037/amp0000151. pubmed.ncbi.nlm.nih.gov/29345485/
  6. Fetters MD, Curry LA, Creswell JW. Achieving integration in mixed methods designs: principles and practices. Health Serv Res. 2013;48(6 Pt 2):2134-2156. doi:10.1111/1475-6773.12117. pubmed.ncbi.nlm.nih.gov/24279835/
  7. Guetterman TC, Fetters MD, Creswell JW. Integrating quantitative and qualitative results in health science mixed methods research through joint displays. Ann Fam Med. 2015;13(6):554-561. doi:10.1370/afm.1865. pubmed.ncbi.nlm.nih.gov/26553895/
  8. Smeallie ET, Choi SW, Mody R, Guetterman TC, Nessle CN. “Better at home”: mixed methods report of intricacies in pediatric febrile neutropenia management. Cancer Med. 2024;13(6):e7106. doi:10.1002/cam4.7106. Open access, CC BY 4.0; figures redrawn and derived numbers computed by Editverse in R. pubmed.ncbi.nlm.nih.gov/38506249/
  9. Corrigendum to “Better at home”: mixed methods report of intricacies in pediatric febrile neutropenia management. Cancer Med. 2024;13(12):e7373. doi:10.1002/cam4.7373. Open access, CC BY 4.0. pubmed.ncbi.nlm.nih.gov/38923352/

Editverse Research Team | Updated October 2026