Describing the research methodology and data analysis processes are paramount for any type of research, as the chosen framework would determine the strength and weaknesses of the particular study. A flawed or incorrect framework could hinder the research efforts and distort the results, eventually leading to a wrong conclusion. This part of the research presents a complete framework for a phenomenological qualitative research design, which was deemed appropriate for exploring the phenomenon of self-perception of new mothers towards their skill in post-partum breastfeeding.
Type of Data
As it was mentioned in the previous sections of this research, the design of this study is qualitative, and its purpose is to collect and analyze subjective qualitative data from the participating young new mothers in regards to their attitude changes towards breastfeeding of the children, as well as the self-perceived levels of efficiency regarding their breastfeeding practices. This data will be collected via open-ended interviews and questionnaires so that the perceptions and phenomenology of the event could be thoroughly studied (Ritchie, Lewis, Nicholls, & Ormston, 2014).
Data Collection Process
The data collection process will involve interviewing new mothers on the subject of breastfeeding. It will consist of the distribution of questionnaires and phone interviews, as well as the processing and analysis of the data. The data will be collected two times – the first set of interviews will take place prior to the proposed intervention in order to measure the current perceptions on the practice of breastfeeding from each individual that had agreed to participate in the research.
The participants would be split into two groups – the test group and the control group. The first data collection process would assist in assigning the women to their respective groups. Optimally, both groups would have an approximately equal number of participants with similar views and experiences of breastfeeding. The second data collection process would happen after the proposed intervention, which would help measure the effectiveness of breastfeeding instructions received via telephone versus the results presented by the control group. Thus, the research design identifies two data collection points – at the beginning and at the end of the research. While it is possible to add a third point of comparison somewhere in between, it will likely be unnecessary (Ritchie et al., 2014).
The estimated length of time for data collection would be one week prior and one week after the proposed intervention, which would continue for a six-week period. There is no point in continuing the intervention for a period longer than six weeks, as by that time it is expected for the majority of mothers to grasp the basics of breastfeeding. Prolonging the procedure would give the test group time to learn these basics on their own, without the support of qualified nurses, thus minimizing the changes in knowledge and attitudes towards breastfeeding between the two groups. One week is an acceptable amount of time to distribute the questionnaires and conduct interviews, as well as receive a response from the participants without putting unnecessary timely constraints both on them and on the researchers (Ritchie et al., 2014).
As the research design for this study is qualitative and phenomenological, an important part of data collection is determining how many samples would constitute data saturation. Unlike with quantitative research methods, there is no clear number on how much is required, and the amount of processed data is usually determined by the convenience and data collection capabilities of the researchers.
However, there are popular estimations for questionnaire types of research on the general attitude towards relatively narrow medical topics. According to Mason (2010), the mean sample size for similar types of studies is 31. As it is within our capabilities to increase that number to 40 samples, we will take that number to ensure data saturation. In order to represent all societal stratum in the scope of this research, an effort would be made to include the members of different economic and racial backgrounds into both participant groups.
Data Quality Enhancement Methods
In order to enhance the quality of data in a qualitative study, the most common and widespread method utilized by the researchers is the triangulation method. This method involves the diversification of the sources and methods used in data collection, in order to provide a more detailed and potentially more accurate picture of the studied phenomenon. In the course of this research, there are ample opportunities for data triangulation. All three methods of data triangulation would be implemented in the following order (Silverman, 2015):
- Source triangulation. The data for this research will be collected from patients with different social and racial backgrounds.
- Method triangulation. While some participants will be asked questions in an interview, others will be required to complete questionnaires containing both open-ended and close-ended questions.
- Analyst triangulation. More than one person will participate in analyzing the data. Several groups of analysts will do so independently of one another and then compare the results.
The data collected in the course of this research will be processed and analyzed two times – the first part of the analysis would happen after the initial results from the first batch of interviews are received. The second analysis would come after the second batch of interviews is completed. The second analysis will also take the results of the first one into account, in order to produce results. Phenomenological data analysis occurs in several stages (Miles, Huberman, & Saldana, 2013):
- All transcripts are read with the purpose of getting a generalized feeling about the responses.
- Second-time reading, which allows breaking the transcript into meaningful parts and compartmentalizing the data into different groups.
- Integrating data groups based on the similarity of content.
- The subjugation of the findings to the free imaginative variation process, which would help determine whether the data units contain a fixed identity and are critical for the phenomena studied in this research (Finlay, 2013).
- Secondary analysis of the raw data in order to ensure it corresponds to the general findings and conclusions of the research.
- Critical analysis of the research methods and design, post-factum.
This data analysis outline is appropriate and traditional for the phenomenological types of qualitative research. It is expected that the proposed data analysis would be efficient at processing the available data and will highlight the major and minor trends in attitudes of young mothers towards breastfeeding before and after the proposed intervention, thus enabling the researchers to form plausible, coherent, and accurate conclusions.
Desired Outcomes for the Data Analysis
There are several desired outcomes for the data analysis, based on the phenomenological qualitative research design. These outcomes are as followed (Silverman, 2015):
- Concrete and detailed descriptions from all participants were appropriately read, analyzed, and compartmentalized based on the information they have provided.
- The phenomenological reduction has been held through the entire process of data analysis.
- Essential meanings and correlations behind the data have been discovered.
- The structure of the studied phenomenon has been articulated.
- The raw data analysis managed to verify or disprove the existing hypotheses.
This study follows a phenomenological qualitative research design. The main data collection tools for the research are in-depth personal interviews and questionnaires containing open-ended and close-ended questions. These tools will be utilized at the beginning and at the end of the research in order to detect any minor or major changes in the participants’ self-perceived attitudes towards personal skills in breastfeeding. In total, the time allocated for data collection is two weeks, spread equally between the start and the end of the active part of the proposed intervention.
The number of participants required to achieve data saturation for this phenomenological qualitative research is forty. In order to ensure maximum data quality and accuracy, triangulation methods will be used to diversify both the sources and the data collection methods for the research.
The data analysis process will involve a thorough and complete investigation, compartmentalization, and integration of the data groups before their subjugation to the free imaginative variation analysis. The expected outcomes of this research include the collection and analysis of all detailed and concrete descriptions from the participants, assessment of the structure of the studied phenomenon, discovery of the essential correlations and meanings behind the data, and reaffirmation of the conclusions made in the course of this research through raw data analysis.
Finlay, L. (2013). Engaging phenomenological analysis. Qualitative Research in Psychology, 11(2), 121-141.
Mason, M. (2010). Sample size and saturation in PhD studies using qualitative interviews. Qualitative Social Research, 11(3), 14-28.
Miles, M. B., Huberman, A. M., & Saldana, J. (2013). Qualitative data analysis (3rd ed.). New York, NY: Sage Publications.
Ritchie, J., Lewis, J., Nicholls, C. M., & Ormston, R. (2014). Qualitative research practice. New York, NY: Sage Publications.
Silverman, D. (2015). Interpreting qualitative data. New York, NY: Sage Publications.