This part of the report gives the reader information about the methods adopted during the gathering of data.
It concludes a section discussing the reliability, validity and generalizability of the research.
Multiple Case Study
This report is presented in the form of a multiple case study. According to Yin (2007), one can adopt three different research strategies within the field of social science. The different strategies include the Exploratory, Descriptive and Explanatory approaches. The authors are in-terested in describing the current leadership situation at three different Team Sportia stores, including the views of both leaders as well as employees. Hence, this research will adopt a descriptive approach. As stated, this report will give a view of the current leadership; therefore, the research is referred to as a cross-sectional study, focusing on the situation at a specific point in time. The results of this multiple case study will hopefully lead to an un-derstanding of leadership and leadership perception that could be used in the development of Team Sportia, similar organizations as well as in future studies.
The method of performing research on two different levels, leaders and employees, is re-ferred to as conducting embedded case studies. The opposite approach to embedded case stu-dies is holistic studies, i.e. only including one level of study, in this case only leaders or only employees (Yin, 2003). Since this report is a multiple case study, the authors have three de-fined research units: Team Sportia Stockholm, Team Sportia Jönköping and Team Sportia Mariestad. Every research unit consists of two levels, which are researched separately: lead-ers and employees. (See figure 1). The units are researched separately but each store is ana-lyzed as a single unit.
Yin (2003) argues that the advantage of conducting a multiple case study, in this case using three different units, is that it is easier to come up with a more relevant analysis. Further-more, the results from a multiple case study are considered more convincing and the results from the study are therefore regarded as more solid (Herriot & Firestone, 1983). However, there are downsides with multiple case studies as well, Yin (2003) argues that the process of conducting a multiple case study is very resource demanding and time consuming that a single student or researcher might have problems managing it.
When conducting a multiple case study, researchers need to select an applicable sample (Yin 2007). Yin (2007) states that when choosing a sample it is important to determine if one will be able to gain the desired information and if it will be possible to get the neces-sary permissions to do so. Regarding this research report, the authors base the sample se-lection on the fact that they have inside information about the chosen samples. Rickard Jo-hansson is a former full-time employee of Team Sportia Mariestad, and is currently part-time employed by Team Sportia Jönköping and Tobias Leander used to be part time em-ployed by Team Sportia Stockholm. The authors believe that the experience gained from working at the three different stores give them an advantage when it comes to gaining and interpreting information. The fact that they know both the current CEOs in person as well as most of the employees is an advantage, since it will enable them to get a more trustwor-thy and accurate picture of the researched environment. However, at all three interview oc-casions all three researches have been present and at least one of which has no previous experience meeting the interviewee, in order to get an as neutral and correct view of the leader as possible. Furthermore, the three chosen samples are somewhat similar when it comes to size and the three CEOs have a similar background, they have all been engaged in the sport retail industry for a long period and they are about the same age. These common traits make the samples interesting from a comparable point of view.
Inductive vs. Deductive Approach
Depending on the nature of a research report, there are two different strategies for collect-ing data, according to Jacobsen (2002). The first strategy, the deductive approach, can be de-scribed as moving from theory towards empiricism. Researchers‟ adapting this strategy starts by making assumptions about reality and then moves on by collecting empirical ma-terial in order to see if the assumptions are aligned with reality. The second strategy, the in-ductive approach, works the other way around. The researcher starts by collecting empirical material almost without any assumptions. Based on the results, new theories are formed (Jacobsen, 2002).
The inductive model could be considered closer to reality than the deductive approach and the two approaches differ when it comes to openness to new information (Jacobsen 2002). Furthermore, Jacobsen (2002) argues that researchers should preferably use a deductive approach when investigating individual‟s view of well-defined conditions. On the other hand, the author argues that the inductive approach is more suitable for vague situations when relevant conditions are hard to predict (Jacobsen 2002).
The authors of this thesis are trying to explain the exercised leadership styles within Team Sportia, through in-depth qualitative interviews with leaders, and one can argue that an in-ductive approach is the more suitable for that purpose. Secondly, since the authors also are aiming at explaining the perception of leadership among employees through quantitative measures, one can argue that a deductive approach should be adopted during this part of the research process. However, the authors argue that the analysis will take a holistic stance, including both the qualitative and quantitative information gathered to create a me-ta-inference and description of the reality researched, hence, a mix of induction and deduc-tion is adapted; see section 3.1.4 for further information.
Mixed Method Design
During this research, the authors adopt a mixed method design, i.e. collecting both qualita-tive and quantitative data. The concurrent embedded strategy in relation to mixed methods is a way of adopting a data collection phase conducting both qualitative and quantitative data simultaneously (Creswell 2009). Further, Creswell (2009) argues that the concurrent em-bedded design uses a primary and a secondary method (qualitative and quantitative) for col-lecting data. The authors of this report are using the quantitative approach as the primary method while the qualitative data is used as the secondary approach, supporting the quan-titative data. When adopting a mixed method, i.e. using both qualitative and quantitative data, it enables the researchers to incorporate and compare the different sources of data, something (Creswell 2009). Creswell (2009) argues that the concurrent embedded strategy could be used when researching different levels or groups and gives an example of this: „if an organization is being studied, employees could be studied quantitatively and managers could be studied qualitatively‟ (Creswell, 2009, p.215). On the other hand, researchers need to be aware that when it comes to analysis of data the concurrent embedded strategy makes it more complicated. The tricky part is that one need to combine the different data col-lected into the analysis in order to make it contributive to the outcome of the study.
According to Jacobsen (2002), there are two different types of data: secondary data and prima-ry data and these are often ideally used together. One needs to be critical when it comes to the use of secondary data; the researcher should evaluate the source and its reliability (Ja-cobsen 2002). Secondary data is referred to data that someone else has gathered earlier for another purpose; while primary data is referred to as data collected for the first time for a the specific research. The authors are gathering secondary data regarding previous research in relation to the study of leadership. This data is used to form an understanding of leader-ship and to introduce the reader to previous research within the field. Furthermore, ten dimensions of leadership are defined by the authors and are used as a foundation for the research of both leaders and employees, i.e. the collection of primary data (See sections 3.2.1 and 3.2.2). Secondary data regarding how to conduct feedback from leaders and em-ployees, which is referred to as 360-degree feedback, is presented together with a section on leadership perception. The selection and use of primary and secondary data described in this research will aid the authors to create an analysis of the researched environment and, hence, answering the research questions and fulfilling the purpose.
Interviews with Leaders
Data are collected from the three different leaders through qualitative interviews. Accord-ing to Klenke (2008), interviews may take different forms, e.g. structured, unstructured or semi-structured. The author argues that in the area of social science, face-to-face interviews are regarded as a corner stone of qualitative research. By using a face-to-face approach, it is possible for the researcher not just to gather information, but also to create a meaning and interpretation of the conversation (Klenke 2008).
During the first part of the interview with leaders the authors are adopting a semi struc-tured interview technique; these questions are formed around the broad concept of leader-ship within Team Sportia (stated further down). When conducting a semi-structured inter-view the interviewers typically prepare a number of questions (Klenke 2008). Klenke (2008) further explains that during the semi-structured interview the interviewers will be able to ask probing questions, such as “can you give an example” and additional questions like “what”
“why” and “how”. The authors are using a uniform body that is used when interviewing all three leaders and depending on the answers, additional questions are formed during the in-terview. Klenke (2008) argues that interviewers may not have the ability to fully understand the interviewees answers and formulate proper questions during the interview due to lack when it comes to knowledge about the researched environment. However, the experience and inside information that the authors possess will be valuable in these situations. Klenke (2008) also mentions generalizability as a disadvantage regarding semi-structured inter-views. The interviews are conducted at each leader‟s home office and a dictating machine is being used to collect the information.
The following twelve pre-established questions are used during all three interviews:
1. Could you please shortly describe your leadership?
2. Do you feel confident in your role as a leader?
3. Which strengths within your organization do you believe will be important for your company’s fu-ture development?
4. What potential sources of development can you identify for you organization?
5. What current trends can you identify within the sport retail business and how do you believe that they will develop during the upcoming five years.
6. Which are your main strengths as a leader according to yourself?
7. Do you argue that you are in control of everything going on within your organization?
8. Where do you turn when looking for advice regarding your leadership?
9. Which channels do you usually use when communicating with your employees?
10. When problems occur in your organization, what is typically the cause? Please give example-/s. (May be either or both internal or external sources of problems).
11. Please tell us about a goal that you really wish to achieve?
12. Finally, is there anything that you would like to add to this conversation?
Surveying of Leaders
During the second part of the interview with the three leaders, the authors are using a structured approach. According to Patton (2002) a structured approach refers to when an interviewer asks each interviewee a number of pre-established questions that has a limited number of response categories. Patton (2002) argues that the backbone of the structured interview lies in the interview protocol, which lists all the questions that are to be explored during the interview.
Some of the advantages of structured interviews include high reliability, easy coding of an-swers, comparable data and low grade of interviewer bias (Klenke, 2008). However, Klenke (2008) argues that there are some disadvantages with this approach as well; when asking close-ended questions (i.e. questions with a set of given answers which the interviewee can chose from) it does not allow the interviewee to express himself/herself in different ways. Klenke further criticizes that a structured approach might have problems of finding out what is relevant to the interviewees (Klenke, 2008). The structured part of the interview in-volves a questionnaire including 35 statements categorized under the ten different concepts (Adaptability, Empowerment, Autocracy vs. Democracy, Stress, Goals, Communication, Risk Propensity, Trust, Consideration and Expertise) related to leadership and previously discussed in the frame of reference part of this report (see section 2.2). The design of the questionnaire was influenced by the concepts of 360-degree feedback, MSP and leadership perception, also previously discussed (section 2).
Previous questionnaires: The authors study the MSP and 360 tools as well as reviewing a bundle of pre-established questionnaires.
Team Sportia: Since this report is a mul – tiple case study within Team Sportia, the authors took the number of employees, the sport retail industry and of course the differences/similarities between the re-search units in consideration.
Dimensions of Leadership: The authors found inspiration while studying the different theories and found questions and statements related to the different dimensions of leadership (described in section 2.2).
The authors pose 35 statements for the managers to reflect on their own leadership, each statement is composed so that the managers can state their response on a scale from 1 to 5. (1 corresponds to “strongly disagree”, 2 equal “disagree”, 3 equal “Neither agree, nor dis-agree”, 4 indicates “agree”, while 5 corresponds to “strongly agree”.) The 5-degree scale is used and recommended when conducting questionnaires, according to Jacobsen (2002).
The questionnaire (Appendix 4) serve as a foundation for the interviews with the managers, the questions are posed during the interview without the managers knowing which dimen-sion of leadership that is discussed. The reason for why the authors do not reveal which dimension each statement is associated with, is to gather as straightforward and honest an-swers as possible. The managers are asked to take a stand with respect to each statement. The statements in the questionnaire are presented with respect to each leadership dimen-sion for the reader to obtain a clearer picture.
Surveying of Employees
The questionnaire used on the leaders is also used when surveying the employees (appendix
5) about their opinion of the exercised leadership. However, the questionnaire is used in a reversed form to suit the role of an employee. Since the same pre-established statements are used for both the employees and for the leaders, the advantages and disadvantages with the named approach are the identical (see 3.2.1). The respondents are asked to reveal whether they are fulltime or part-time employed at Team Sportia, other than that the res-pondents are ensured total anonymity. In order to get an as accurate and unbiased result as possible the survey is sent to the employees by mail and the respondents are asked to send their answers back to the authors in a pre-paid envelope. This way the employees will be given the opportunity to respond to the statements at home, taking their time and not be-ing biased by their leader. The questionnaire, which is sent to the employees, does not re-veal which leadership dimension each statement is associated with, however the authors will present the survey with respect to each leadership dimension for the reader to obtain a clearer picture.
Essential to keep in mind when analyzing qualitative data are three different concepts: First: Description, referring to the process of sorting and giving a clear view of the collected data. Second: Systematization, the process of simplifying the view of the data. Finally: Combi-nation, the creation of meaning and interpretation of the data (Jacobsen, 2002). The authors of this report are summarizing the qualitative data (raw data from interviews using a dictat-ing machine) collected through the interviews with leaders, interpreting it and the result is presented in sections 4.1-4.3. Note that this section contains a subjective interpretation and analysis of the data gathered, i.e. the presentation of the data may differ depending on who compiles it. The data is sorted under sub-headings to simplify for the reader. Jacobsen (2002) suggests that quotes preferably could be used to create additional meaning of the da-ta.
Regarding the quantitative data that is collected through surveying of employees (added to this: the leaders‟ response to these statements), the authors will present the raw data in ap-pendix 6, but also summarize and present this information more lucidly in the analysis sec-tion. In contrast to the qualitative analysis, the presentation of the quantitative data does not include any interpretations by the authors, i.e. the results would be presented in the ex-act same manner regardless of who compiles it. Jacobsen (2002) argues that there are three types of answering alternatives: categorical, ranking and metric. For this research the authors are using the ranking alternative, since the respondent is asked to give an answer ranging from one to five on a ranking scale.
The statistical measures available for presenting the central tendency of data that are re-trieved from ranking answers include Mode and Median. Mode is the most frequent answer while median is the center of the distribution of answers (Jacobsen, 2002). In this thesis the authors use mode as a primary measurement for central tendency, and support their analy-sis with investigating the median answer if appropriate. Further Jacobsen explains that the measures used to describe the spread of ranking answers are: Range, Min/Max and Modal Percentage. Range refers to the difference between the highest and the lowest response, min/max is the highest and lowest response in absolute terms and modal percentage can be explained as the mode divided by total number of respondents to a specific question (Jacobsen, 2002).
In order for the authors to test whether the questionnaire used to research the perceived leadership among employees in this report is reliable, a reliability test is conducted. A reliabil-ity test can also be called a test-retest reliability, which refers to the fact that, ceteris paribus, if handing out a questionnaire to a person at two different times you should get the same res-ponses. A tool, named Cronbach’s alpha, has been developed in order to test reliability (Field, 2009). Further, Field (2009) explains that when running a reliability test and analyzing the Cronbach alpha output, one should accept values that are exceeding 0.7, indicating a relia-ble test, and reject values lower than 0.7, regarded as not reliable. Reliability tests should be performed on different factors of a questionnaire (Field, 2009), in the case of this report it implies that the authors will perform clustered tests on statements related to each of the ten different dimension of leadership described in section 2.2. The authors will have to reverse the scale on statements 2, 4, 7, 12, 14 and 27, in order to run the reliability analysis in SPSS. The reason for this is that the statements are reverse phrased. Take statement 4 for exam-ple: a high score on statement 4 indicates low level of empowerment while a high score on statement 5 indicates a high level of empowerment. Hence, the responses to statement 4 will be reversed (1 equals strongly agree, 5 equals strongly disagree and so forth) in order to be able to run the test. This reversal of responses will only occur in the reliability analysis and one-way ANOVA analysis; it does not apply to any of the other analyses performed in this report. Using reverse-phrased statements is a way for researchers to reduce biased res-ponses among participants, since they force them to read every question or statement care-fully (Field, 2009). The results of the reliability test are presented in appendix 7.
Table of Contents
1.2 Company Overview – Team Sportia
1.3 Industry Overview – The Sport Retail Sector in Sweden
1.4 Problem Area
2 Frame of Reference
2.2 Dimensions of Leadership
2.3 360-degree Feedback
2.4 Perception of Leadership
3.1 Research Approach
3.2 Data Collection
3.3 Data Analysis
3.4 Critique of Method
4 Empirical Material
4.1 Gren – Stockholm
4.2 Liedholm – Mariestad
4.3 Nordahl – Jönköping
4.4 Questionnaire Employees
5.4 One-way ANOVA Analysis
5.5 Final Analytical Observations
8 List of References
GET THE COMPLETE PROJECT