Is there an optimal number of people and names to include
in a case study?
Fanny L. Sandberg
Halmstad University
This article investigates the importance of the number of partakers in a case study. By interrogating the average number of people included in case studies across different fields, a speculation of the numbers impact can be presented. The article explains the negative sides of having too many people included and aims to provide the reader with understanding for the sincerity of having just the right number of names.
A case study is an applicable method when it comes to a whole range of different fields. Unlike many types of studies a case study is constructed to bring out the details from the viewpoint of the participants by using multiple sources of data (Tellis, 1997). This makes it an ideal methodology when an extensive investigation is needed (Feagin, Orum & Sjöberg, 1991). These tools can be applicable as a general theory in many different fields including anthropology, business, communications, economics, education, medicine, political science, technology, social work and sociology (Gerring, 2007).
It exists a conventional view of case studies that claims that a case study cannot be of value by itself (Flyvbjerg, 2003).
Dogan and Pélassy (1990, p.121) expresses it like this: “one can validly explain a
particular case only on the basis of general hypotheses. All the rest is uncontrollable, and so of no use.” Since the case study method is an example of a qualitative research, the critic directed against it affects the applicability on other industries (Baxter & Jack, 2008). To make a science a good science it needs to have a high validity and a high reliability (Taylor &
Small, 2002) and the narrowness of a case study may lower both factors. This view has improved over the years. By viewing further research on the subject, it comes to knowledge that many scientists have changed their perception about case studies (Flyvbjerg, 2003). Donald Campbell (1975) is one of these examples. He used to be a tough resistant to the case study method but now explains that his work has gone through a major change. He desbribes his latest work like this (Campbell, 1975):
“After all, man is, in his ordinary way, a very competent knower, and qualitative common-sense knowing is not replaced by quantitative knowing… This is not to say that such common sense naturalistic observation is objective, dependable, or unbiased. But it is all that we have. It is the only route to knowledge-noisy, fallible, and biased though it be” (p. 191)
This will on the other hand not be the focus of this article. And even though some previous research has been done about case studies this paper attempts to make one contribution to the literature. Based on a combination of the previous studies and a quantitative study presented in this paper a discussion about the optimal number of names and persons in a case study will come forth.
Purpose of a Case Study The purpose of a case study is to create a depiction of a certain company or event.
This makes it possible to study a complex phenomenon within its context (Baxter &
Jack, 2008). A case study can inform professional practise or evidence-informed decision making in both clinical and policy realms even though it is a form of qualitative research (Baxter & Jack, 2008).
According to Yin (2003) it can support the deconstruction and the subsequent reconstruction of various phenomena by enabling the exploration of individuals or organizations through complex interventions, relationships etc. Further, Yin (2003) describes that a case study approach should be considered when: (a) the focus of the study is to answer “how”
and “why” questions; (b) you cannot manipulate the behaviour of those involved in the study; (c) you want to cover contextual conditions because you believe
they are relevant to the phenomenon under study; or (d) the boundaries are not clear between the phenomenon and context. To make a case study as realistic as possible the sample in the case should represent the total population (BMJ, 2015).
Genders in Case Studies The business case for gender equality becomes stronger and stronger. This rises the expectation for equal representation divided on the different genders (de Vries, 2015). Even though improving the balance between men and women in all academic ranks have been on many universities and governments agendas for quite some time (e.g. Rees, 2001; Bailyn, 2003; Özbilgin &
Healy, 2004), the equality has not yet reached its top (van den Brink, Benschop
& Jansen, 2010). Case studies are a way to increase the transparency of an organisation. This organisational opacity could lead to a more beneficial situation for women since gender bias are more likely to occur when procedures and operations are kept confidential (e.g. Ledwith &
Manfredi, 2000).
Reliability and Validity in a qualitative study
The expressions reliability and validity are mainly connected to quantitative research (Golafshani, 2003) and could therefore seem unnecessary to discuss whilst discussing a qualitative study method. Golafshani (2003) did on the other hand extend current theories and redefined the meaning of reliability and validity in order to make them applicable for qualitative studies as well.
Joppe (2000) defines reliability as:
“[…]the extent to which results are consistent over time and an accurate
representation of the total population under study is referred to as reliability and if the results of a study can be reproduced under a similar methodology, then the research instrument is considered to be reliable.” (p.597)
According to Charles (1995) the reliability could be determined through the test-retest method at two different times.
This attribute of the instrument is referred to as stability. If the measure is stable, then the results should be similar. Some critic has been directed towards this method, mainly by Joppe (2000). Her point is that the test-retest method may create an awareness for the respondent to the subject matter, and therefore influence the responses given. Thus, Crocker and Algina (1986) say, “Test developers have a responsibility of demonstrating the reliability of scores from their tests”.
(p.106).
When it comes to validity, the traditional criteria are founded in a positivist tradition where positivism has been described by a methodical theory of validity. Joppe (2000) defines this in a quantitative research perspective:
“Validity determines whether the research truly measures that which it was intended to measure or how truthful the research results are. In other words, does the research instrument allow you to hit
‘the bull’s eye’ of your research object?
Researchers generally determine validity by asking a series of questions, and will often look for the answers in the research of others” (p.106).
Wainer and Braun (1998) defines validity as the initial determination of which data is to be gathered and how this is supposed to be done.
Even though these factors are mainly associated with quantitative studies Patton (2001) accent the awareness for people constructing qualitative studies as well.
Stenbacka (2001) build the prior foundation for the need to redefine these terms in order to fully be able to apply them on qualitative studies. She describes the concept of reliability as one of the quality concepts in qualitative research which “to be solved in order to claim a study as part of proper research” (p.551). Davies and Dodd (2002) argue that the application of the core of the concept in qualitative research should differ from those in quantitative research by “accepting that there is a quantitative bias in the concept of rigor, we now move on to develop our reconception of rigor by exploring subjectivity, reflexivity, and the social interaction of interviewing” (p.281).
Methodology
To do this research, case study methods from different fields have been examined and evaluated. The fields have been narrowed down to four main ones, with five cases in each field. The metrics that have been studied are the following:
1. Number of names and persons included
2. Professions /Titles of these persons 3. The diversion between the genders To make it possible to draw extra conclusions, the different fields have also been included. The diversion between the genders have been presented with the percentage of women included. This percentage have been rounded up to
numbers with one decimal. All the calculations in the empirical section have been calculated with SPSS and Microsoft Excel. The collection of data has been displayed and processed as a quantitative
study. The framework is summarized in Table 1.
Table 1
A collection of data from different Case Studies
Case number
Number of names included
Titles included %
women
Field
1 7 Doctor, director, associate, advisor,
president
28,5 Medicine
2 4 director, provincial governor, regional chief, director-general
0 Medicine
3 2 Vice president and president of company 0 Arts
4 13 manager, founder, competitor, marketing
director, HR director, potential employees
38,5 Marketing
5 12 field directors, CEO, entrepreneur, head of department, chief, manager, vice president
33,3 Business
6 4 CEO, intern, manager, purchaser 0 Business
7 5 CEO, engineer, owner, previous employees 20 Business
8 8 Volunteer workers, doctors 25 Medicin
9 12 Minister of health, parlament member,
medical student, embassadeur, health promotion specialist, doctor
25 Medicin
10 8 soprano, AD, administration, financial
controller, head of marketing, HRM director
50 Arts
11 2 executive manager, priest 0 Business
12 1 business owner 0 Marketing
13 2 lawyer, accountant 0 Marketing
14 5 managing director, director, buyers,
manager, business partner (2kv)
40 Business
15 6 account inspector, CEO, vice president, general manager
0 Marketing
16 6 AD, CEO, head of department 0 Arts
17 2 Managing directors 0 Marketing
18 6 director, founder, doctor 0 Medicine
19 6 vice president, associate, marketing VP, marketing director, AD, managing director
0 Arts
20 4 senior manager, managing director,
marketing manager
25 Arts
The average of this sample has been calculated as the following:
( ∑Names included ) /20 = 115/20 = 5,75
Calculating the average is a tool to describe a population or a sample. The main idea is to provide the reader with a fast overview of a certain situation (Undersökningsguiden, 2016). The average value could sometimes result in misleading indications. To avoid this, a calculation of the standard deviation has been included.
s= the standard deviation of a sample x= each value in the data set x̅= mean of all values in the data set
n= number of data points
s= 3,49247311≈ 3,5
A limitation of this study is the number of case studies evaluated. By examining 20 different case studies an idea can be
created but the sample is too small to provide an overall picture of the total population.
Considering the usual gender diversion this has also been evaluated in the study.
The total number of women included are 24. Counting the percentage of the total amount of names:
24/115=0,2086…≈ 21%.
A histogram may not be the best approach when dealing with a rather small sample (Laerd Statistics, 2013).
Considering this, a normality line has been left out, and the purpose of the graph is to present the type values. A type value defines what the most frequent value is (Ericsson et al., 2011). In this study the type values are 2 & 6.
Results
There are a few points noted in the attached table. All these points have been located under the chosen themes from the theory part. The first one is the minority of women included. The research based
Figure 1
on a sample proves a female rate of barely 21%. Many of the cases had a total lack of females and therefore only male characters.
The female gender is underrepresented in some industries (e.g. Dynan et al., 1997;
Ellemers et al., 2004; Paxton et al., 2012).
To avoid this bias, a range of different fields were included in the study. Viewing the different positions, it becomes clear that the professional representation spreads among the top positions of the companies. This gives the impression that some important employees have been left out. A case study is a qualitative form of research (Baxter &
Jack, 2008). This makes the need for a perfect representation is bigger. The gender bias could also be a result of the general underrepresentation of women on top positions. This is another reason to examine
the different characters. Even though not all hierarchic levels are necessary for the main story line, someone below vice president could be included to make the cases more reliable. If the study is fictional the implementation and rationing of certain groups are more practicable. If the study is based on a real situation and some groups are underrepresented an evaluation of the cases credibility should be considered. Is this an applicable case to all situations this company might encounter?
The amount of people included is another factor to be considered. Reading different cases gives a good impression of the effect of the number of names and persons included. Some of the cases studied had only 1 or 2 persons in the story. Fewer characters are easier to keep track of. On the downside, it decreases the anticipation and makes the reader lose interest more quickly.
By lengthening the case, more characters can be included. There should theoretically be a positive correlation between the page numbers and the amount of characters.
Some of the cases processed have had up to 13 different characters to keep track of.
Those cases make the story harder to follow and the case study loses its point. The important matter is to not draw focus from the actual story line of the case. Since it can be tricky to find the balance with the combination of necessities and the things that keep the story interesting, some extra time should be spent doing this. Based on this empirical study the magical number seems to be between 3 and up to 6 names included.
When it comes to the validity and reliability of the case study methodology, the theory of Stenbacka (2001) have been implemented. Stenbacka applied these terms on qualitative studies which makes
them relevant to study in this situation.
Together with Davies and Dodd (2002) the view of validity and reliability from a qualitive perspective was developed. The important note to have in mind while evaluating a qualitative method from this perspective is to accept the existence of a quantitative bias. The point of this article is that even though the quantitative bias should be accepted, it could be reduced though by thinking the case through. A higher level of consistency between the sample and the population would result in a decreased quantitative bias.
Conclusion
The purpose of this paper is to describe the importance of the number of people and names included in a case study. The main process, outlined in Table 1, has features which range from the gender division to the field in question. The limitations in this study may have affected the results which decrease the empirical validity.
There are a few different aspects that affects the perception of a case when doing a case study. Given this research paper, the number of names and persons included in the case is a major one. With too many characters included the context gets blurry and the story gets lost. But having too few persons gives the impression that important employees are missing from the cases. I have been going through a large number of cases. One of the criteria fulfilled for a case study to be good is that it needs to be applicable on other situations. By underrepresenting some groups, women for example, the strong feeling that it misses a quantitative touch decreases.
To summarize; to build a good case for a case study these points should be considered:
1. Include a minimum of 3 and maximum of 6 characters.
2. Make sure the gender diversion corresponds with the reality.
3. Look over the professions mentioned, are all the necessary people mentioned?
4. Don’t steal attention from the story.
References
Bailyn, L. (2003). Academic Careers and Gender Equity: Lessons Learned from MIT. Gender, Work And
Organization, 10(2), 137-150.
Baxter, P. & Jack, S. (2008). Qualitative Case Study Methodology: Study Design and Implementation for Novice
Researchers. The Qualitative Report, 13(4), 544-559. Retrieved from http://www.nova.edu/ssss/QR/QR13- 4/baxter.pdf [Accessed 2016-11-22]
Campbell, D. (1975). "Degrees of Freedom" and the Case Study.
Comparative Political Studies, 8(2), 179, 191.
Charles, C. (1995). Introduction to educational research (1st ed.). White Plains, N.Y.: Longman Publishers USA.
Crocker, L. & Algina, J.
(1986). Introduction to classical and modern test theory (1st ed., p. 106). New York: Holt, Rinehart, and Winston.
de Vries, J. (2015). Champions of gender equality: female and male executives as leaders of gender change. Equality, Diversity And Inclusion: An International Journal, 34(1), 21-36.
http://dx.doi.org/10.1108/edi-05-2013- 0031
Dogan, M. & Pélassy, D. (1984). How to compare nations (1st ed., p. 121).
Chatham, N.J.: Chatham House Publishers.
Dynan, K. & Rouse, C. (1997). The Underrepresentation of Women in Economics: A Study of Undergraduate Economics Students. The Journal Of Economic Education, 28(4), 350-368.
http://dx.doi.org/10.1080/0022048970959 7939
Ellemers, N., Heuvel, H., Gilder, D., Maass, A., & Bonvini, A. (2004). The underrepresentation of women in science:
Differential commitment or the queen bee syndrome?. British Journal Of Social Psychology, 43(3), 315-338.
http://dx.doi.org/10.1348/0144666042037 999
Gerring, J. (2007). Case study research (1st ed., p. 2). New York:
Cambridge University Press.
Golafshani, N. (2003). Understanding Reliability and Validity in Qualitative Research. The Qualitative Report, 8(4), 597-606. Retrieved from
http://nsuworks.nova.edu/tqr/vol8/iss4/6 [Accessed 2016-11-25]
Feagin, J., Orum, A., & Sjoberg, G.
(1991). A Case for the case study (1st ed., pp. 27-79) Chapel Hill: University of North Carolina Press. 27-79
Flyvbjerg, B. (2003). Fem missförstånd om fallstudieforskning. Statsvetenskaplig Tidskrift, 106(3), 185-206.
Joppe, M. (2000) The Research Process, The Quantitative Report Journal, 8(4), pp.597-607. Retrieved from http://www.nova.edu/ssss/QR/QR8- 4/golafshani.pdf [Accessed 2016-11-25]
Ledwith, S. & Manfredi, S. (2000),
Balancing gender in higher education, The European Journal of Women’s
Studies,7(1)
Patton, M. Q. (2002). Qualitative evaluation and research methods (3rd
ed.). Thousand Oaks, CA: Sage Publications, Inc.
Paxton, M., Figdor, C., & Tiberius, V.
(2012). Quantifying the Gender Gap: An Empirical Study of the
Underrepresentation of Women in Philosophy. Hypatia, 27(4), 949-957.
http://dx.doi.org/10.1111/j.1527- 2001.2012.01306x
Rees, T. (2001). Mainstreaming Gender Equality in Science in the European Union: The ‘ETAN Report’. Gender And Education, 13(3), 243-260.
http://dx.doi.org/10.1080/0954025012006 3544
Stenbacka, C. (2001). Qualitative research requires quality concepts of its own.
Management Decision, 39(7), 551-555
Taylor, P. & Small, B. (2002). Asking applicants what they would do versus what they did do: A meta-analytic comparison of situational and past behaviour employment interview
questions. Journal Of Occupational And Organizational Psychology, 75(3), 277- 294.
Tellis, W. M. (1997). Application of a Case Study Methodology. The Qualitative Report, 3(3), 1-19. Retrieved from
http://nsuworks.nova.edu/tqr/vol3/iss3/1 [Accessed 2016-11-19]
van den Brink, M., Benschop, Y., &
Jansen, W. (2010). Transparency in Academic Recruitment: A Problematic Tool for Gender Equality?. Organization Studies, 31(11), 1459-1483.
http://dx.doi.org/10.1177/0170840610380 812
Wainer, H. & Braun, H. (1988). Test validity (1st ed.). Hillsdale, N.J.: L.
Erlbaum Associates.
Yin, R. (2003). Case study research (1st ed. p. 118). Thousand Oaks, Calif.: Sage Publications.
Özbilgin, M. & Healy, G., (2004) ‘The gendered nature of career development of university professors: The case of
Turkey’. Journal of Vocational Behavior 64(2), 358–371
Websites
BMJ Publishing Group, Populations and samples | The BMJ. (2016). Bmj.com.
Retrieved from
http://www.bmj.com/about-bmj/resources- readers/publications/statistics-square- one/3-populations-and-samples [Accessed 2016-11-23]
Ericsson, E., Nyholm, T., Hagman, H., Carselind, J., & Janghede, S.
(2011). Lägesmått |
Matteguiden. Matteguiden. Retrieved from http://www.matteguiden.se/matte- a/statistik/lagesmatt/ [Accessed 2016-11- 26]
Gunnarsson, R.
(2002). Standardavvikelse. Infovoice.se.
Retrieved 26 November 2016, from http://www.infovoice.se/fou/bok/1000004 9.shtml [Accessed 2016-11-26]
Spridningsmått. (2016)
Undersökningsguiden. Retrieved from http://www.undersokningsguiden.se/sprid ningsmatt/ [Accessed 2016-11-26]
Testing for Normality using SPSS Statistics when you have only one independent variable.
(2013). Statistics.laerd.com. Retrieved from https://statistics.laerd.com/spss- tutorials/testing-for-normality-using-spss- statistics.php [Accessed 2016-11-26]
Case studies
1. Sullivan, E., Drobac, P., Thompson, K., &
Rodriguez, W., (2011). Botswana’s Program for Preventing Mother-to-Child HIV
Transmission, Boston: Harvard Business Publishing [Case study]
2. Charumilind, S., Jain, S., & Rhatigan, J., (2011) HIV in Thailand: The 100% Condom Program. Boston: Harvard Business
Publishing [Case study]
3. Cecillon, JF., Saintilan, P., (2013) Did you find the voice of God in the data? Melbourne:
COLLARTS Australian College of the Arts.
[Case study]
4. Prill, E., (2009) Taking a Page from AbeBooks: The Evolution of a Team, Cambridge: Institute of Technology. [Case study]
5. Russo, M., & Goldstein, D., (2007) Seventh generation: Balancing customer expectations with supply chain realities. Oregon: Lundquist College of Business, University of Oregon [Award winning case study]
6. Gill, A., McKay, G., (2008) Inland Glass &
Aluminium Ltd. and Aluminum Curtainwall Systems Inc. British Columbia Innovation Council [Case study]
7. Huber, S., & Søilen, K., (2006) Mercatus Engineering, Lund: Studentlitteratur AB [Case study]
8. Dhillon, R., & Rhatigan, J., (2011) The Measles Initiative, Boston: Harvard Business Publishing [Case study]
9. Bitton, A., Talbot, J., Clarke, L & Kadarfor, E., (2011) Tobacco Control in South Africa, Boston: Harvard Business Publishing [Case study]
10. Cecillon, JF., Saintilan, P., Cannon, R., &
Smellie, M (2013) Crisis in the opera house,
Melbourne: COLLARTS Australian College of the Arts [Case study]
11. Barinaga, E., Älverbrandt, L., (2014) Origo Resurs, Copenhagen: Copenhagen Business School [Case study].
12. Gracia, J., (2000), Sundae Sales get Sweeter, Watertown: Give to get Marketing [Case study]
13. Spencer, I., (1991), Cape Breton Chalets, Antigonish: St. Francis Xavier University [Case study]
14. Wright, A., & Follows, S., (1995), Mountain Gap Resort, Wolfville: Acadia University [Case study]
15. Mukund, A., (2002), HR Restructuring - The Coca Cola and Dabur way, India: IBS Center for Management Research [Case study]
16. Cecillon, JF., Saintilan, P., Cannon, R., &
Smellie, M (2013) The taste Case, Melbourne:
COLLARTS Australian College of the Arts.
[Case study]
17. Mukund, A., (2001), HR Restructuring - Kellogg’s Indian Experience, India: IBS Center for Management Research [Case study]
18. May, M., Rhatigan, J., & Cash, R., (2011), BRAC’s Tuberculosis Program: Pioneering Dots Treatment for TB in Rural Bangladesh, Global Health Delivery Project. Boston:
Harvard University [Case study]
19. Saintilan, P., (2013), The Artistic v.
Marketing Interfunctional Conflict Case, Melbourne: COLLARTS Australian College of the Arts. [Case Study]
20. Saintlan, P., & Smellie, M., (2013), The International Meeting, Melbourne:
COLLARTS Australian College of the Arts.
[Case study]
Fanny L. Sandberg is a student at Halmstad University. She studies International Marketing, with six months left to her Bachelor’s degree in Business.
I appreciate the helpful comments from my fellow student Linnea Czarnecki while doing this research. I also benefitted from previous research on the subject and the instructions from prof. Klaus Solberg Søilen.
.
