Introduction: Defining the Quality of Research Organization
In the qualitative study carried out by Karin Egberg Thyme, Britt Wiberg, Berit Lundman and Ulla Hällgren Graneheim and titled Qualitative content analysis in art psychotherapy research: Concepts, procedures, and measures to reveal the latent meaning in pictures and the words attached to the photographs (Thyme, Wiberg, Lundman & Graneheim, 2013), the target population has been identified quite precisely as “women with depression” (Thyme et al., 2013, p. 103). The accessible population has also been defined in a rather careful manner; the research mentions “a woman in her early thirties diagnosed with depression according to DSM-IV” (Thyme et al., 2013, p. 103). Seeing that the choice of the research participant was quite sporadic, it can be assumed that the sampling probability method was used in the course of the research.
As far as the sampling method is concerned, the authors refer to it as the “sessions of PDT/Scribbling” (Thyme et al., 2013, p. 103). Nevertheless, it can be assumed that the researchers have used one of the probability types of qualitative research sampling methods known as the simple random method. Indeed, it is not stated in the study whether the research participant was selected based on a specific set of characteristics; it is assumed, though, that the participant fits the characteristics introduced as the key parameters for the research subjects. It should also be kept in mind that the size of the sample population (one person) seems rather small for academic research. For the study to have the required objectivity rate, it is imperative that at least two people should be included in the study. More to the point, little to no sampling biases are identified in the study, and no information concerning the dropout has been provided.
Assessing the Measurement Instruments
The researchers have provided sufficient information on the data collection methods. According to what the authors of the paper claim, the information was retrieved with the help of the so-called “PDT/Scribbling” (Thyme et al., 2013, p. 103). The required data was retrieved by recording the visual and the spoken information acquired from the research participant. Apart from the records of the process of data retrieval, the visual information included scribbles, drawings, and notes that the research participant made. The data was recorded by two of the researchers.
As far as the measurement is concerned, it should be mentioned that the authors of the study, in fact, research the possibilities for qualitative data measurement. Consequently, the measurement methods represented in the paper can be viewed as unique and even somewhat questionable. For example, the researchers offer their study participant a mixed test, therefore, making it clear that the measurement tool will also combine the elements of several methods. To be more exact, the test results delivered by the participant are measured by the so-called “two-dimensional model” (Thyme et al., 2013, p. 102) with the help of a “qualitative content analysis” (Thyme et al., 2013, p. 102), with the horizontal axis allowing for measure the data epistemologically, and the vertical axis providing the chance for utilizing “the methodological approach from close and concrete descriptions to distant and abstract interpretations” (Thyme et al., 2013, p. 102). The level, at which the research variables are measured, seems quite legitimate. As far as the data collection instrument is concerned, the above-mentioned PDT/Scribbling should be brought up again. A pilot study, however, was not conducted prior to writing the research under analysis.
Data Collection and the Veracity of the Information
As it has been specified above, the data collection process is in the limelight of the given study. Defined as the “qualitative content analysis” (Thyme et al., 2013, p. 102), it presupposes that the research participant is provided with an opportunity to engage in art therapy and produce specific “artworks” (e.g., drawings, painting, all kinds of scribbles, etc.), which will later on undergo the procedure of coding and decoding, thus, providing the researchers with a specific pattern, which will later on be tested.
It is essential that, apart from the visual information, the participant also provides the researchers with audio materials. Indeed, according to the record of the process of research, both the images were drawn by the participant and her responses have been incorporated into the study, with every single sound made by the subject of the research having been recorded carefully for the further analysis.
It is also quite remarkable that the authors of the study suggested analyzing not only the elements of the [participant’s speech that had a certain semantic load, but also those that did not seem to make any sense when taken outside of their context (e.g., sighs, coughs, interjections, etc.). By incorporating both verbal and non-verbal messages into the study, the researchers have made it possible to reduce the amount of the study limitations. For example, the chances for misinterpreting the response of the participant due to its ambiguity has been reduced greatly, with the provision of such an important source of information on the participant’s response as the elements of non-verbal communication, including the gestures made by the study participant, the facial expressions, the changes in the participant’s mood over the course of the experiment, etc.
Speaking of the reliability of the instrument used in the research, the idea of incorporating a series of visual tests is not new and, therefore, has been tested in a range of other studies. According to the studies conducted previously, the instruments in question, i.e., the qualitative content analysis and the two-dimensional model, which are utilized as the key tools for results measurement, have been used in research for quite a while. In fact, qualitative data analysis, though being generic, is often considered “one of many qualitative methods used to analyze textual data” (Forman & Damschroder, 2008, p. 40). Therefore, having been time tested several times, the method in question can be considered quite trustworthy and efficient in determining the implications of the outcomes of qualitative research. Though the method specified above has been around for only thirty years (Mayring, 2000), it has already warranted its active use in modern qualitative research.
As far as the second instrument is concerned, the two-dimensional model is also used in scholarly researches quite frequently as a tool for measuring qualitative data. As researchers explain, the use of the two-dimensional model allows for a comprehensive analysis of a specific phenomenon, with the consideration of both the theoretical aspects of the phenomenon in question and its practical implications: “One dimension delineates three nonlinear core processes of information seeking activities: opening, orientation, and consolidation. The other dimension consists of three levels of contextual interaction: cognitive approach, internal context, and external context” (Zhang & Wildemuth, n. d., p. 10). Though no pilot study was conducted prior to the research, the method chosen for the analysis of the problem that the researchers deal with in their study is quite valid.
Critiquing the Descriptive Statistics: What Can Be Improved
Since the research is focused on checking the viability of the qualitative content analysis as a research tool by conducting an experiment involving the assessment of the visual and audio material provided by the research participant, it can be assumed that the efficacy of the qualitative content analysis is the dependent variable, whereas various types of data provided by the research participant can be viewed as the independent variables.
Most of the types of descriptive statistics are represented in the paper. However, it is quite remarkable that, among the types that have been included in the analysis, there are no measures of central tendency. On the one hand, such a choice of data analysis can be explained by the fact that the research only had a single participant; on the other hand, the lack of the measures of central tendency does somehow devalue the significance of the results. On the other hand, the lack of statistical information regarding the results of the research makes one question the validity of these results, as well as the credibility of the evaluation of the results.
Likewise, the measures of variability have been omitted in the study. Such a choice of descriptive statistics for the research can be justified by the fact that the study does not have many numerical data to offer. Being a qualitative study, it does focus on the interpretation of the results and the processes of coding and decoding the data much more than it does on the statistical analysis of the acquired information. The measures of shape and size, however, have been included into the study, though they have not been introduced via numbers – the authors of the research preferred to split the acquired information into codes and categories in accordance with the data coding method of analysis. Much to the authors’ credit, though, the research results arranged in tables with clear sections, which make the research results very easy to understand. The data in the tables coincides with the information offered in the body of the article fully.
Critiquing the Inferential Statistics: Making Forecasts
Though the research features little to no statistical data, inferential information is provided in the study. Owing to the process of coding and decoding the responses obtained from the research participant, as well as the further interpretation of these responses with the help of the two-factor model mentioned above, the premises for making assumptions concerning the efficacy of the qualitative content analysis as a means of considering qualitative data are created (Thyme et al., 2013). The research results are provided both in the tables and in the discussion section. While the tables offer only a general idea of the outcomes of the study, the discussion of the experiment provides a full view of what tasks have been carried out and why the qualitative content analysis can be considered a valid tool for qualitative research. Though the paper has minor issues with the definition of central tendency, it can be considered a rather compelling study.
Reference List
Forman, J. & Damschroder, L. (2008). Qualitative content analysis. Empirical methods for bioethics: A primer (pp. 39–62). San Diego, CA: Elsevier.
Mayring, P. (2000). Qualitative content analysis. Forum: Qualitative Social Research, 1(2), para. 1–7. Web.
Thyme, K. E., Wiberg, B., Lundman, B. L. & Graneheim, U. H. (2013). Qualitative content analysis in art psychotherapy research: Concepts, procedures, and measures to reveal the latent meaning in pictures and the words attached to the pictures. The Arts in Psychotherapy, 40(1), 101–107.
Zhang, Y. & Wildemuth, B. M. (n. d.). Qualitative analysis of content. Web.