Focus Groups
When studying a specific phenomenon such as the role of critical success factors in the change management process, the qualitative method of focus group discussion can be very useful. Focus groups usually allow scholars to gain insight into the thinking process of the participants in the study to have a profound understanding of the topic under investigation. In contrast to surveys, which limit the feedback of respondents to the answers to specific questions, focus groups can be used to capture and collect more insightful information with greater efficiency and economy. The economic aspect is especially noteworthy because most studies require scholars to collect the necessary information without wasting resources.
Group interaction is at the core of focus group research because the way that group members communicate usually encourages the revelation of different ideas, connections, and discussions. A skilled facilitator is essential when collecting data through the use of focus groups due to the need to capture not only verbal but also nonverbal communication. Importantly, participants involved in a focus group may have different responses to the topic of discussion, especially regarding sensitive themes.
When exploring the issue of change management and the factors that contribute to its success, the experience and expertise that the different people in a focus group can provide are the most valuable takeaway from this type of discussion.
Despite the challenges associated with sensitive ideas and topics, focus groups present a useful research tool in qualitative studies because each member of a selected group has unique characteristics related to a study subject. Therefore, the possibility is high that group members will influence each other’s ideas by responding to questions that might never have been brought out in a regular setting.
This method of data collection is the most effective in a situation where a scholar must discern the attitudes of participants towards a new proposal or program, identify positive and negative sides, assess whether the process under consideration is working, and evaluate potential success in a specific setting. Thus, focus groups fit the current study as it aims to determine which success factors are necessary to guarantee the effectiveness of a change management process in healthcare organizations.
On the downside, in using focus groups to collect relevant information, researchers risk encountering the issue of groupthink. When people are brought together in a group to discuss a specific problem or phenomenon, a common trend involves the leading and influential members affecting the decisions and opinions of the other people taking part. This influence presents a threat to the reliability of the findings because some participants who may have differing opinions on a subject will fail to express their thoughts to counter a dominant idea that prevails within a group. Also, some participants may be reluctant to express negative attitudes in a group setting when face-to-face with other people. Indirect research, in comparison, offers an opportunity for participants to feel freer and more open to express negative attitudes.
In explaining the methodology of focus groups, several components should be considered. To start, exploration is the primary objective of focus groups, suiting the need to learn about an issue of importance – in this case, success factors in the change management process – from the target population. After identifying the purpose of the focus group, it is then essential to explore conceptualization and logistics, which are essential methodology components of a focus group. The definition of the study sample and the population comprises the next step.
Sample Population
The population represents the group of people who will participate in a focus group to be analyzed by a researcher. The sample represents a section of the identified population. To recruit participants, it is necessary to employ a sampling method. Probability sampling, while considered the most beneficial and bias-free sampling method, is usually difficult to implement in the case of focus groups. In any research effort, although it is always desirable to use a representative sample that incorporates the various aspects of an entire population and includes as many types of subjects as possible, in focus group research that concentrates on a narrowed topic area, random sampling would be costly and time-consuming.
Non-probability convenience sampling is the most likely solution to this issue because of the need to make sure that a sufficient number of participants will be involved as well as that they will be accessible to the scholar and will not require expending many resources. In this type of sampling, each member of the target population meets the inclusion criteria formulated in advance. Moreover, each must be accessible to the researcher, in close geographic proximity, available at a specific time, and willing to participate.
Convenience samples are often referred to as accidental because they are usually selected in the sample ‘simply as they just happen to be situated, spatially or administratively, near to where the researcher is conducting the data collection’ (Etikan, Musa & Alkassim 2016, p. 2). Overall, the chosen sampling method is affordable, easy, and quick to implement, in addition to being readily available to the researcher.
Since convenience sampling is a method that allows researchers to find participants who are nearby and are available, the sample for the focus group in this study will be recruited in healthcare facilities situated in the local area. The researcher will visit several facilities and ask nurses, nursing managers, and other personnel about their willingness to participate. A brief of the study with its objectives, purpose, and research questions will be handed out to potential participants along with the researcher’s contact information. A week will be given for follow-up, and if not enough participants are willing to give their time, the researcher will visit other local facilities to ensure the choice of a viable number of group members.
A sample of ten participants will be sufficient for the current study. This size is optimal for promoting discussion among participants and making sure that the facilitator keeps up with the group. It is always recommended to keep the number of participants lower than expected due to the need to allow each person to speak and share his or her insights (Krueger & Casey 2015). The researcher will develop a list of open-ended questions in the development stage of the study (DeFranzo 2014).
Five questions are usually the optimal number when it comes to focusing groups because this quantity is manageable and will allow participants to discuss essential topics without diverging from the main theme of the group session. While open-ended questions imply less control, they are more beneficial in the context of focus groups because they promote discussion.
Facilitation of Focus Groups
Facilitation is another important step in conducting focus groups, involving such components as preparation, the pre-session stage, and the focus group session itself. During the stage of preparation, the researcher will prepare himself or herself for the focus group session. It is expected that a focus group team, consisting of two participants, will be developed. One will act as a facilitator while the second will take notes to aid in the data collection process. A note-taker is essential to focus groups because of the need to record feedback from participants. Also, making the facilitator responsible for taking notes has the potential to impede the conversation due to the likely need for the facilitator to stop and record pertinent information.
In the pre-session stage, the facilitator will have the opportunity to familiarise himself or herself with group dynamics. Important takeaways from this step include taking note of participants who are shy and thus hold back from interacting with other group members. To ensure that these group members will engage in the conversation, the facilitator can designate an area in the room where all participants will introduce themselves, after which the facilitator will encourage everyone to take their chairs and gather around that area. Making participants comfortable by engaging in small talk before the session is also important.
When the facilitator is welcoming to the group, the likelihood of group participants engaging in a fruitful conversation about the study subject will be higher. Facilitation itself implies following the script formulated earlier in the process. If the group session is to be recorded, the facilitator will inform the members of that fact. Also, it is important to remind participants that all information will be kept confidential if it is true. After opening the session, the facilitator will proceed with the questions. At the end of the session, the researcher will have a transcript to be used for further data analysis.
Data Analysis: Thematic Analysis
In the course of the focus group session, the researcher will develop a collection of important notes about the discussion about success factors in the change management process. Also, the full recording of the focus group session will be completed to make sure that all information is included in the analysis. The analysis of relevant data will take place immediately after the group session ends. Because of this, the researcher must ensure that the notes taken are comprehensive and include as much detail as possible. The data will be analyzed with the help of thematic analysis, which is used in analyzing qualitative data through the development of themes and patterns that relate to the studied topic at hand.
According to Alhojailan (2012), ‘thematic analysis is considered the most appropriate for any study that seeks to discover using interpretations. It provides a systematic element to data analysis […] and allows the researcher to associate the analysis of the frequency of a theme with one of the whole content’ (p. 2). To identify relevant themes within the process of data analysis, the scholar will develop a concept map that helps to summarize the main points of the gathered information. The concept map will depict the main question topics and the themes that emerged for each of the topics. This will make the process of thematic analysis easier as well as goal-oriented.
One key advantage of thematic analysis is that the method is flexible in terms of theory, and it provides an opportunity to adjust the process of data analysis based on the researcher’s needs. Thematic analysis can be beneficial for answering questions about participants’ experiences and feelings while also revealing how well people understand a particular idea or perceive a phenomenon. Also, scholars can use a variety of ways to approach thematic analysis.
On the one hand, it is possible to use induction – developing themes and codes based on the gathered data. On the other hand, scholars can use deduction – developing codes and themes as a result of initially formulated ideas and concepts.
The thematic analysis used in the current research will consist of six phases that include becoming familiar with the data, generating familiar codes, looking for relevant themes, reviewing those themes, defining them, and completing a write-up. After the notes are made and after the recorded information is transformed into a transcript of the focus group session, the researcher will familiarise himself- or herself with the transcript and generate codes by systematically organizing information.
Open coding is the most appropriate solution in the case of the current research because discussions that result from focus group questions cannot be predicted, and thus, the researcher may need to modify codes to match them to the transcript. This coding method aligns with the inductive approach, which focuses on theme development based on data. Also, open coding is associated with the grounded theory framework of attaching codes and concepts to observed data (Foley & Timonen 2015).
The next step in data analysis will imply searching for themes that represent significant and important aspects of data. At the stage of searching for themes, the codes will be examined closely to identify whether they could be fitted together to form a theme. For example, if several participants were to indicate that collaboration was the most necessary factor to achieve success in the change management process, an appropriate theme to underline this phenomenon should be developed.
Also, searching for themes comprises the step of thematic analysis that requires collating data for each potential theme and making sure that they are reviewed for viability. At the end of the stage of searching for themes, the codes will be organized into broad themes, each of which will be dedicated to a specific topic about the research question that the scholar will have formulated in the beginning.
After all the themes have been identified, the researcher will review them. All data relevant to all themes will be gathered together to group them into themes. The use of data analysis software will help the process take less time. Colour-coding will aid the researcher in grouping data into themes, with each theme assigned a designated color classification. The next stage is defining themes to explain the essence of each. The researcher will answer such questions as ‘what is the theme saying?’ ‘how do the themes relate to each other?’ ‘if there are any subthemes, how well do they interact with the main themes?’ and so on. The last step in thematic data analysis is the write-up. The results of the analysis will be presented in the ‘Findings’ section of the research.
Questionnaire
Questionnaires represent a data collection instrument regularly used in both qualitative and quantitative research but for different purposes. These are developed from a set of standardized questions (items) organized in a specific way to obtain information about different topics. The key difference between questionnaires and interviews lies in the fact that the former takes place in a unified fashion, which means that all respondents have the opportunity to answer the same questions.
Questionnaires provide a cheap, efficient, and quick way for a scholar to collect large volumes of data from any size sample of study participants. This approach is beneficial because when participants answer questions in this form, the researcher does not need to be present to guide the entire process. With focus groups, in comparison, the researcher plays the role of a facilitator and is responsible for making sure that each study question is discussed and that all relevant notes are taken for use in data analysis. Important aspects to consider in the case of designing a questionnaire include the length, the order of questions, and the presentation.
For example, potential respondents usually find long questionnaires discouraging, especially when they do not have enough time to complete all the questions. Also, questions should be organized logically, from the least to the most sensitive or from general to specific, to ensure that some answers will not be influenced by the previous ones. Lastly, the researcher should ensure that the language of the questions is clear and concise.
When using questionnaires in research, a scholar should take into account both the advantages and disadvantages of this research tool. On the one hand, questionnaires allow collecting feedback on a particular phenomenon from a large sample of participants, especially when it is impractical to use a range of different intensive methods (The University of Sheffield 2014). Questionnaires developed based on a particular structure offer scholars the opportunity to explore trends and patterns that describe the characteristics of an idea as well as identify what is occurring in the studied context. Thus, the benefits of questionnaires include a quick collection of data, the opportunity to obtain feedback from a large sample population, and the ability for participants to give anonymous answers to encourage honesty and openness along with the capacity for rapid processing of the data with the help of software.
On the other hand, this method presents a risk of respondents interpreting questions differently. This points to the need for scholars to design questionnaires in such a way that minimizes the likelihood of inadequate interpretation. Also, questionnaires usually depend on the sampling frame chosen by the researcher, which can be challenging in the event of difficulty in identifying a relevant or accurate sampling frame. Lastly, there is an issue of motivating respondents to complete their questionnaires, which means that the researcher will have to follow up with participants on the phone or via e-mail to make sure that all questionnaires are completed.
In the present study, the researcher will use the standardized method of a 10-point rating scale, also known as the Likert scale. By developing a 10-point rating scale, a researcher asks respondents to quantify their perceptions based on the questions presented by the questionnaire. This type of questioning is predominantly used to measure attitudes and opinions with a wider scale of detail than regular ‘yes/no’ questions may allow.
However, 10-point Likert scales have recently been criticized in the research literature because of their inability to offer a true mid-point. As a result, the researcher decided to offer a 0 (zero) option to survey respondents, giving the scale in this study 11 points in total. Also, it was mentioned that the ‘magic zero’ had a unique effect on questionnaire respondents – the zero is intuitively perceived as the lowest and understandable rating, which means less confusion among respondents (Primary Intelligence 2017). The Likert scaling method in questionnaires was thus chosen to uncover different degrees of opinion and better understand the feedback of the participants.
Testing for Reliability and Validity
Reliability shows the consistency or dependability of the measurements used in tests or experiments. For example, if a research participant takes the same test multiple times, in terms of reliability, the researcher expects obtaining the same results each time. Validity occurs when there is a clear recognition that a test measures the phenomena that it was designed to measure. Testing both the validity and reliability of a questionnaire will allow the researcher to ensure that the developed questions are effective for answering what they were intended to answer.
To measure reliability, the test-retest method can be used. This is linked to the extent to which participants’ scores are consistent throughout a given period. To assess test-retest reliability, it is necessary to measure the results of a group of participants at one time and then repeat the same measurement at a later time. The goal of the test-retest is to find a correlation between the two sets of scores (Price, Jhangiani & Chiang 2015). In the case of questionnaires, completing the test-retest method will allow the researcher to ensure that the respondents give reliable answers as well as provide respondents the opportunity to reconsider their attitudes towards the studied phenomena.
Validity is yet another vital aspect to take into account because of the risk of measurements being reliable without actual validity, a challenge to a researcher. An absurd example of this issue was provided by Price, Jhangiani, and Chiang (2015), who recommended imagining a person believing that an individual’s self-esteem is directly correlated to the length of his or her index finger. The expected consequence would be to try to measure the self-worth of other people by holding a ruler up to their index fingers. This example shows that good test-retest reliability can have no validity whatsoever.
In the case of the current research and questionnaires, the scholar will measure content validity to identify the level to which a measurement covers the studied topic. By the conceptual definition, content validity is associated with attitudes defined by feelings, thoughts, and actions towards a particular idea or phenomenon. Therefore, it is expected that each respondent who will have completed questionnaires will experience either a positive or negative attitude about the exercise. To reach a good level of content validity, the researcher will ask respondents to give their feedback on the exercise in a way that reflects their actions, feelings, and thoughts. The measurement itself is qualitative and will be assessed by conducting a short interview with participants.
Sample Population
Choosing an appropriate sample size to address the needs and the requirements of the research is essential for ensuring statistical significance. The minimum sample size will be determined with the help of a power analysis, which is a technique used for determining the probability of a statistical analysis being able to reach a false null hypothesis. In other words, the sample size is statistically significant when there is no possibility of making a Type-II error (Filho et al. 2013). Even though a larger sample size means a higher statistical power of the analysis, a very large sample size when using questionnaires will involve high costs in terms of resources, time, and effort.
Thus, the key to achieving a statistically significant sample size in the study is a generalization. This involves extending results obtained from a smaller sample size of study participants to the whole population.
Given the study scope, it is expected that a sample size comprising thirty questionnaire respondents is a viable sample that the researcher can access. Because it is expected that the researcher will have to hand out questionnaires to potential participants in person and avoid using e-mails, it is important to consider time and effort in terms of resources. However, the sample of participants should be representative of the entire population to extend the obtained results, meaning that random sampling should be used to reduce bias and enhance representation.
Selection of Participants
In the methodology section dedicated to the discussion of focus groups, convenience sampling was chosen to facilitate the researcher’s process of recruiting participants. It is important to note that focus groups are harder to manage, meaning that the scholar had to prioritize convenience over randomization to make sure enough participants would be recruited for the group session. In the case of the questionnaire, a random sampling method will be applied to enhance representativeness and decrease the possibility of bias. In the search for an appropriate sampling method, simple random sampling was chosen as the tool to recruit respondents for the questionnaire.
This method is associated with working with a population in which an individual has an equal opportunity of being selected. An example of this method would be choosing several random names from a list of two hundred employees in an organization. Every employee on the list has an equal chance of being chosen because no system or framework governs which persons will have their names selected from the list.
For the questionnaire in this study, the method of using a list of employees will also be applied. A list of healthcare facilities in the area local to the researcher will be compiled to randomly select the facility from which potential participants will be recruited. Each facility will be assigned a number to make the process of selection easier. With the help of randomization software, a number will be generated, and the facility assigned that number will be chosen for a further sampling of participants.
The researcher will then contact the facility’s management and ask for a list of employees. If the management refuses to provide a list or says that they are not interested in allowing their workers to participate in a study, the randomization software will generate a new number, assigned to another facility, which the researcher will contact. Importantly, the researcher will guarantee confidentiality so that no personal information will be exposed to the general public. Using the same method, each employee in the list will be given a number, and the randomization software will generate thirty random numbers. Then, the researcher will contact each individual to ensure that he or she will agree to participate.
In the case that some of the thirty selected employees do not wish to participate, the researcher will further use the randomization software to generate new numbers and contact other participants to reach the desired sample size. After all potential participants are contacted, the researcher will negotiate a convenient time to meet with them and give them the printed copies of the questionnaire. Respondents will be given a week to fill out and return the questionnaire to the researcher. In the case of any emergencies, the participants will contact the researcher and negotiate an appropriate time frame to complete the survey.
The manual handling of questionnaires was chosen over web-based surveys because it offered the researcher an opportunity to communicate with participants and explain the study objectives as well as obtain some brief feedback. Also, when a questionnaire is sent via e-mail, the response rate tends to be lower due to a lack of commitment as well as technical issues (Davern 2013).
Data Analysis: Chi-Square Test
Since the questionnaires will deal with Likert scales to measure the extent of respondents’ agreement–disagreement or satisfaction–dissatisfaction with a statement, statistical analysis should cater to their specific nature. From the variety of methods that may be used for analyzing Likert scale data, the chi-square statistic method was chosen for this research. This implies comparing participants’ actual responses to questions with expected answers. In the context of the current study, the researcher will compare the responses of each participant to each question. The wider the deviation between participants’ responses, the higher the chi-square statistic will be. The level of deviation will show how well the results of the analysis fit the hypothesis developed at the beginning of the study.
Both the chi-square test for independence and the chi-square goodness of fit will be performed during the data analysis of the questionnaires. The researcher will combine participants’ responses into distinct categories to make the process easier. For example, responses such as ‘strongly disagree’, ‘slightly disagree’, and ‘disagree’ may be grouped within a ‘disagree’ category. Then, a chi-square test will be run using an Excel spreadsheet program that enhances data organization and management.
The next step is to examine the results that the program generated. The primary focus involves paying closer attention to the level of statistical significance and the size of the chi-square statistic (Hall 2018). A higher chi-square statistic will indicate a wider variation between participants’ responses. A lower statistic, on the other hand, will show the researcher that respondents have similar attitudes to a phenomenon or statement.
The interpretation of the results will simply identify the presence of a relationship that is statistically significant and exists without the revelation of any data on the strength of those relations. A chi-square test is a good option for the present study because of the possibility of respondents leaving some questions unanswered. To address this issue, the chi-square test will allow the researcher to add a ‘no answer’ category. While an analysis that involves the comparison of participants’ answers allows examining the relationships of the data, it is also possible to identify non-response sampling errors as well as their relationships.
Reference List
Alhojailan, M I 2012, ‘Thematic analysis: a critical review of its process and evaluation’, West East Journal of Social Sciences, vol. 1, no. 1, pp. 39-47.
Davern, M 2013, ‘Nonresponse rates are a problematic indicator of nonresponse bias in survey research’, Health Services Research, vol. 48, no. 3, pp. 905-912.
DeFranzo, S E 2014, Advantages and disadvantages of open questions in course evaluations. Web.
Etikan, I, Musa, S A & Alkassim, R S 2015, ‘Comparison of convenience sampling and purposive sampling’, American Journal of Theoretical and Applied Statistics, vol. 5, no. 1, pp. 1-4.
Filho, D B F, Paranhos, R, da Rocha, E C, Batista, M, da Silva Jr., J A, Santos M L W & Marino, J G 2013, ‘When is statistical significance not significant?’, Brazilian Political Science Review, vol. 7, no. 1, pp. 31-55.
Foley, G & Timonen, V 2015, ‘Using grounded theory method to capture and analyse health care experiences’, Health Services Research, vol. 50, no. 4, pp. 1195-1210.
Hall, S 2018, How to use a chi square test in Likert scales. Web.
Krueger, R A & Casey, M A 2015, Focus groups: a practical guide for applied research, 5th edn, SAGE Publications, London.
Price, P C, Jhangiani, R S & Chiang, I-C 2015, Research methods in psychology, 2nd edn, BCampus, Toronto, Ontario.
Primary Intelligence 2017, The magic in a 0-to-10 rating scale. Web.
The University of Sheffield 2014, Questionnaires. Web.