Considerations on the use of IRAMUTEQ software for qualitative data analysis

Dear Editor,

It was because of my interest in papers that address software-based qualitative data processing that I became a reader of this journal. Therefore, I have carefully analyzed the article called “The use of IRAMUTEQ software for data analysis in qualitative research”¹.

Considering the use of IRAMUTEQ, I highlight the following aspects:

1. in the Method, the transcription and the size of the corpus do not interfere in the variation in the number of lines of the Text Segments (TS) in IRAMUTEQ, which is, on average, 3.25 lines^{2), (3}.

2. the nomenclatures of Initial Context Unit (ICU) and Elementary Context Unit (ECU) are from the Alceste software. In IRAMUTEQ, the respective terms are corpus and Text Segments^{2), (3}.

3. in Results, IRAMUTEQ does not perform encodings, since they are all done by the researcher as a prerequisite. Furthermore, the researcher cannot manipulate the data, so performing “separation of information” does not correspond to the advantages of using IRAMUTEQ^{2), (3}.

4. in Results, interpretation is not a step for the conduction of DHC in IRAMUTEQ.

5. in Results, Figure 1, in IRAMUTEQ the corpus is not divided into subcorpus in the production of classes. As it is presented, there is no analysis of the corpus, but of the subcorpus. However, the subcorpus is(are) not presented.

6. in Results, the total of Text Segments (“964 ECU”) is far below the expected value for a corpus of 21 semi-structured interviews of approximately 40 minutes each⁴.

7. in Results, classes are formed according to the relationships between the Text Segments that result from the decomposition of the corpus^{2), (3), (4), (5}.

Finally, there are several distorted interpretations in the use of IRAMUTEQ, which lead to discussions and conclusions that should be amended.

REFERENCES

• ¹
  Souza MAR, Wall ML, Thuler ACMC, Lowen IMV, Peres AM. The use of IRAMUTEQ software for data analysis in qualitative research. Rev Esc Enferm USP. 2018;52:e03353. DOI: http://dx.doi.org/10.1590/S1980-220X2017015003353
  » https://doi.org/http://dx.doi.org/10.1590/S1980-220X2017015003353
• ²
  Camargo BV, Justo AM. Tutorial para uso do software de análise textual IRAMUTEQ [Internet]. Porto Alegre: Universidade Federal de Santa Catarina; 2013 [citado 2015 out. 15]. Disponível em: http://www.iramuteq.org/documentation/fichiers/tutoriel-en-portugais
  » http://www.iramuteq.org/documentation/fichiers/tutoriel-en-portugais
• ³
  Camargo BV, Justo AM. IRAMUTEQ: um software gratuito para análise de dados textuais. Temas Psicol [Internet]. 2013 [citado 2015 out. 15];21(2):513-8. Disponível em: http://pepsic.bvsalud.org/pdf/tp/v21n2/v21n2a16.pdf
  » http://pepsic.bvsalud.org/pdf/tp/v21n2/v21n2a16.pdf
• ⁴
  Ramos MG, Lima VMR, Amaral-Rosa MP. IRAMUTEQ Software and discursive textual analysis: interpretive possibilities. In: Costa A, Reis L, Moreira A, editors. Computer Supported Qualitative Research: WCQR 2018. Philadelphia: Springer; 2019. p.58-72.
• ⁵
  Ratinaud P. IRAMUTEQ: interface de R pour les analyses multidimensionnelles de textes et de questionnaires (computer software) [Internet]. 2009 [cited 2015 Oct 15]. Available from: http://www.iramuteq.org
  » http://www.iramuteq.org

ANSWER LETTER

Authorship SCIMAGO INSTITUTIONS RANKINGS

Dear Editor,

We would like to thank you for the opportunity to answer the questions from a reader of your renowned journal about “The use of IRAMUTEQ software for data analysis in qualitative research”, published in volume 52 of 2018. We answer each question below, as well as inform the article and tutorial that we used as primary references to write about the specific use of the software in our research.

1. Most of the times, Text Segments (TS) are approximately three lines long, scaled by the software according to the size of the corpus. However, the researcher can set up the division of the text segments. For example, in cases where there is a large number of short answers to an open-ended question from a questionnaire, it is advised that each text be defined as a single TS, and the software is set up not to segment it^{1), (2), (3}. The passage that addresses the transcription refers to the constitution of the corpus, which is done by the researcher.

2. In the IRAMUTEQ^{1), (2} tutorial and in the article “IRAMUTEQ: free software for textual data analysis”⁴, on which the research was based, the aforementioned terms were used as advised by the authors. However, it is known that in the tutorial of the November 2018 version these terms were suppressed. Therefore, these nomenclatures are no longer used³.

3. The encoding is indeed done by the researcher, but the program also requires encoding. This is an advantage for the author to identify the verbatim from the interviews, which also makes interpretation and analysis easier. It is true that the researcher should not manipulate data, however, the separation by text segments is done in classes, which facilitates interpretation.

4. The interpretation is not a step, but the researcher’s analysis.

5. Below we identify the search base of the IRAMUTEQ TUTORIAL where that item can be found¹:

In the DHC tab of the results, one can access the dendogram, which shows the partitions that were made in the corpus until the final classes were obtained. The dendogram should be read from left to right. In the example of Figure 11, at first the “obesity” corpus, for instance, was divided (1st partition or iteration) into two subcorpus. Secondly, a subcorpus was divided into two (2nd partition or iteration) and thus we obtained class 5. Thirdly, there are more partitions, originating, on the one hand, classes 1 and 2, and on the other, classes 3 and 4. The DHC stopped here because the 5 classes were stable, that is, composed of units of text segments with similar vocabulary. In addition to the dendogram, this result interface enables us to identify the lexical content of each of the classes (to access it, click on the Profils tab) and a factorial representation of the DHC (to access it, click on the AFC tab). In the November 2018³ tutorial, the term used is subcorpora.

• vi) We used 90.56% of the corpus. In the Camargo tutorial^{1), (2}, including that of Camargo³, the percentage used was 75%. In the tutorial, there is an example of corpus that obtained 994 segments with 81% use. We followed the acceptance guidelines suggested by the authors.

• vii) The specification regarding the creation of classes and their writing can be done to evidence several aspects, according to the purpose of the article, as long as in compliance with the tutorial.

The preparation of the article, submitted to this Journal on April/1/2017 and published in October/2018, as well as the related analyses, were made according to the researcher’s references regarding the results. We believe that the readers’ input can strengthen and increase the reliability of any research, promoting its growth and strict rigor in the investigation. That is very enriching for us as researchers. We are at your disposal to clarify our references and methodological conducts at any time.

Sincerely,

The authors:

Marli Aparecida Rocha de SouzaMarilene Loewen WallAndrea Cristina de Morais Chaves ThulerIngrid Margareth Voth LowenAida Maris Peres

REFERENCES

• ¹
  Camargo BV, Justo AM. Tutorial para uso do software de análise textual IRAMUTEQ [Internet]. Porto Alegre: UFSC; 2013 [citado 2019 fev. 12]. Disponível em: http://www.iramuteq.org/documentation/fichiers/tutoriel-en-portugais
  » http://www.iramuteq.org/documentation/fichiers/tutoriel-en-portugais
• ²
  Camargo BV, Justo AM. Tutorial para uso do software IRAMUTEQ [Internet]. Porto Alegre: UFSC; 2016 [citado 2019 fev. 12]. Disponível em: http://www.iramuteq.org/documentation/fichiers/Tutorial%20IRaMuTeQ%20em%20portugues_17.03.2016.pdf
  » http://www.iramuteq.org/documentation/fichiers/Tutorial%20IRaMuTeQ%20em%20portugues_17.03.2016.pdf
• ³
  Camargo BV, Justo AM. Tutorial para uso do software IRAMUTEQ [Internet]. Porto Alegre: UFSC; 2018 [citado 2019 fev. 12]. Disponível em: http://iramuteq.org/documentation/fichiers/tutoriel-portugais-22-11-2018
  » http://iramuteq.org/documentation/fichiers/tutoriel-portugais-22-11-2018
• ⁴
  Camargo BV, Justo AM. IRAMUTEQ: um software gratuito para análise de dados textuais. Temas Psicol [Internet]. 2013 [citado 2019 fev. 12];21(2):513-8. Disponível em: http://pepsic.bvsalud.org/scielo.php?script=sci_arttext&pid=S1413-389X2013000200016&lng=pt http://dx.doi.org/10.9788/TP2013.2-16
  » https://doi.org/http://dx.doi.org/10.9788/TP2013.2-16» http://pepsic.bvsalud.org/scielo.php?script=sci_arttext&pid=S1413-389X2013000200016&lng=pt

Publication Dates

• Publication in this collection
  23 May 2019
• Date of issue
  2019