qual quan

FIELD METHODS López-Sintas / CONTINGENCY TABLE García-Álvarez, Contingency Table: A Two-Way Bridge between Qualitative ...

0 downloads 69 Views 281KB Size
FIELD METHODS López-Sintas / CONTINGENCY TABLE García-Álvarez,

Contingency Table: A Two-Way Bridge between Qualitative and Quantitative Methods ERCILIA GARCÍA-ÁLVAREZ IESE, Universidad de Navarra

JORDI LÓPEZ-SINTAS Universidad Autónoma de Barcelona Combining qualitative and quantitative research methods in the same project allowed the authors to gain more in-depth knowledge as well as to achieve sharper data reduction and verification than would have been possible using only one approach. Qualitative data were obtained from in-depth interviews with thirteen founders of leading family businesses and published secondary data. The authors used qualitative computer software to codify founders’ values from the full transcriptions of their interviews. The resulting contingency table was the bridge to visually explore the relationship between codes (values) and primary documents (family firm’s founders) using multidimensional scaling. The authors then used hierarchical clustering to form a taxonomy of founders. Finally, they returned to qualitative analysis and studied the association between founders’ values and family background for external validation of the taxonomy.

The literature in the field of management recognizes the important role of values in family businesses (Dyer 1986; Ward 1987). Recent work has focused on identifying values that are common to family businesses (Corbetta 1999) and attempts (1) to understand how those values differ from the values of nonfamily businesses (Ward 1999) and (2) to isolate the values that can contribute to success (Cappuyns, Gallo, and Vilaseca 1998). As far as we know, however, typologies of entrepreneurs are not commonly based on values (La Fuente, Salas, and Pérez 1985; Donckels and Fröhlich 1991). We present here the mix of qualitative and quantitative methods we used in studying the values of founders and in developing a typology of those values. The contingency table is a two-way bridge that allowed us to integrate quantitative exploratory techniques in the qualitative analysis to gain more in-depth knowledge and sharper data reduction. The process we describe is highly iterative and linked tightly to data.

Field Methods, Vol. 14, No. 3, August 2002 270–287 © 2002 Sage Publications

270

García-Álvarez, López-Sintas / CONTINGENCY TABLE

271

GETTING STARTED Before starting fieldwork, we did a literature review to focus our research problem and to develop a theoretical framework that would guide our research design and subsequent analysis. We identified constructs for each of the four main components of our conceptual framework—founder (fifteen constructs), family (twenty), firm (twenty-three), and environment (four)— and used these constructs initially as a guideline to protocol development and later to help us in our codification process. Our goals focused on the following: • determining the founders’ value system, • finding out if founders in family business could be differentiated according to their value systems, and • identifying any relationship between founders’ values, as well as any differences in these values and the founders’ personal background.

SELECTING CASES We focused our research on founders of family businesses who are approaching retirement age (sixty-five years old) and currently working with the potential successors of their firms (i.e., founders who have extensive business experience in sound firms and who are on the verge of transferring power from one generation to another). We chose this group for two reasons: (1) The majority of family businesses in Spain are between the first and second generation and have not experienced succession (Gallo and García Pont 1989). This is one of the most critical issues in family business survival. (2) The study of founders and their family businesses at this stage allows us to link entrepreneurship to the family business field by analyzing the initial business concept and motivations of each founder when they set up their business and the unique characteristics of each organization. One methodological problem involved in studying family businesses is building a sample of firms that are owned and managed by one or several families. These firms are not always easy to identify. We used two steps in building our sample: (1) We identified first-generation family businesses and their founders from “Las primeras 500 empresas de Galicia” (The top five hundred businesses of Galicia) (Gómez and Martínez 1992) and (2) we verified that these firms were still family run in 1996 by checking several business directories.

272

FIELD METHODS

We obtained a final list of twenty-eight founders who owned firms of varying sizes with all the required characteristics; this was our theoretical sample (Glasser and Strauss 1967). We sent these founders a personal letter and later phoned them to make an appointment for an interview. This was very difficult as most had never before agreed to be interviewed. In some cases, we made more than twenty phone calls to convince members of our sample to participate in the study. In the end, thirteen founders agreed to participate once we guaranteed confidentiality.

THE INTERVIEWS We wanted to obtain the founders’ own point of view, so we used an in-depth, semistructured interview to produce extensive texts. To develop a final interview protocol that ensured we could cover the main themes from our conceptual framework in all cases, we developed a matrix that linked each question to the information we expected to obtain from the answers and to the constructs from our conceptual framework. This is a two-dimensional matrix, in which the rows are the information we expected to obtain (themes) and the columns are the protocol questions. Each cell contains the initial codes derived from our theoretical framework. We developed an interview protocol, then, that covered all relevant pieces of our previous theoretical framework, allowed us to test the quality of information obtained after each question, and facilitated the coding process. In developing our interview protocol, we were also careful with the language and terms we used. We wanted to formulate questions with everyday language and avoid business jargon. This, we felt, would make founders feel more confident and comfortable, eliminate bias due to the interviewer’s introducing terms, and allow respondents to introduce specialized terms on their own. This, in turn, would let us identify differences across cases in language and terms used. We were also careful in the sequencing of questions. We began with a description and then asked questions about experience, behavior, opinions, and values. We included several questions to verify information, looking for contradictions (Patton 1990, chap. 7). To avoid founders’ possible reluctance to discuss sensitive topics (such as their succession), we posed a question that caused the interviewee to assume the possibility of his own death: “How do you see this firm in . . . years?” We varied the number of years in this question according to the founder’s age so that he clearly had no chance to be a future participant. All founders answered the question without difficulty and began

García-Álvarez, López-Sintas / CONTINGENCY TABLE

273

to talk openly about the possible consequences of their own death for both their family and their firm. As a result, we obtained the information we were interested in without pressing the interviewee and while still keeping the interview relaxed and pleasant. All interviews were conducted by García-Álvarez and took place in each founder’s office, lasting from two to six hours. Each interview was tape recorded. Both authors handled the transcription and analysis of the thirteen interviews to ensure thorough knowledge of the process.

COMPUTER-AIDED QUALITATIVE SOFTWARE We chose Atlas.ti for text management (Muhr 1997) because of its flexibility. This software allowed us to maintain close contact with our text data throughout the data analysis and to extend our analysis to summaries and memoranda.

ANALYSIS We pursued an iterative analysis in three main steps. (1) First, we carried out an in-depth, case-by-case analysis, coding for themes (using Atlas.ti) until we derived the final code book. This produced a founders-by-values matrix. (2) Next, we applied quantitative techniques of matrix analysis, using SPSS, to look for and display graphical patterns in the coded data. (Atlas.ti converts case-by-attribute matrices to SPSS-compatible data matrices. This feature facilitates combining qualitative and quantitative techniques in data analysis.) (3) Finally, we developed a qualitative backup of our cross-case patterns. The complete procedure is summarized in Figure 1. We explain each of these steps further in the following sections. Other authors have combined qualitative and quantitative techniques but with somewhat different research designs; see, for instance, Monge and Eisenberg (1987), Rice and Danowski (1993), Stohl (1993), Jang and Barnett (1995), Jehn and Doucet (1996, 1997), and Doerfel and Barnett (1999).

WITHIN-CASE DATA: CODE BOOK DEVELOPMENT After transcribing the thirteen interviews, we segmented them into relevant pieces of text of varying lengths (quotations) and began descriptive cod-

274

FIELD METHODS

FIGURE 1 Screen Shot of Questionnaire Design Window in Entryware Designer Software 3.0 1- Full transcripts of interviews 2- Adaptation of transcripts to work with Atlas.ti 3- Creation of textual quotations 4- Revision

ATLAS.ti

5- Descriptive coding 6- Revision

TEXT. & CONCEPT. ANALYSIS

7- Descriptive code reduction 8- Revision

Codebook

9- Conceptual code reduction, and code families 10- Revision

Contingency table

11- Networks 12- Revision 13- Case taxonomy 14- Conceptual networks display

SPSS: MDS & Ward's algorithm

15- Textual evidences from networks 16- Founders' types and variables summary matrix of variables 17- Coherence matrix between groups of founders and their behavior Discussion and conclusions

ATLAS.ti CONCEPTUAL NETWORKS MATRIX ANALYSIS

ing. We began with a previous code list obtained from our interview matrix and the literature review. We augmented the theme list as we coded more interviews, using both open coding (a new code is assigned to a segment of text) and in-vivo coding (a word from the text is used as a new code; see Bernard and Ryan 1998:611–19; Strauss and Corbin 1998). Then, we began descriptive code reduction by merging codes and reviewing code grouping and code meaning. We entered memos throughout the process to record the development of the ideas and concepts as they emerged during the coding stage. After several revisions of the accuracy of our conceptual codes, we produced our final code book. Then, we grouped our codes into three main families: the main components of our conceptual framework (founder, family, firm, and values), relevant information from our protocol matrix (successor’s socialization, potential successors), and aspects that emerged during analysis (development of skills and acquisition of resources, founder’s resource origin).

García-Álvarez, López-Sintas / CONTINGENCY TABLE

275

FIGURE 2 Founders’ Relationships ~Founders' formal education

~Family background -

~Family Background + ~Factor -

~Business experience

~Hard childhood

~Founders' mentor ~Start-up cause

~Activity background -

~Activity background +

CF:Background

Founder

~Stability

CF:Founders' values

~Innovation

~Growth ~Satisfaction

~Economic interest

~Ambition ~Hard work

~Negative human relations ~Negative failure perception ~External control

~Risk aversion

~Prosperous life

~Positive failure perception

~Seriousness ~Honesty

~Task orientation ~Family sense

~Positive human relations

~Constancy

~Simplicity

~Risk-taking

~Internal control

~Entrepreneurship ~Active life ~Autonomy

~Sacrifice ~Rigor

~People orientation

~Determination

~Short-term orientation

~Gratitude

~Long-term orientation

~Ethical orientation

~Cosmopolitan orientation

~Altruism ~Local orientation ~Family orientation

~Sense of Guilt ~Vocation

~Sense of achievement

~Business orientation

We found it useful to build a graphic display of the associations among themes (this is a feature of Atlas.ti), a network of the major theme-level and family-level codes that we obtained during our analysis. (Graphic displays of this kind are helpful in organizing codes obtained during analysis.) Figure 2 is the graphic display related to the founder, his values, and social background. One of our main concerns was to control subjectivity as much as possible. The method we used was for both of us to independently follow the same analytic procedure (i.e., we would each get output, compare our level of agreement, discuss our discrepancies, and reformulate the output, case by case with the whole sample). The revision steps in our analytic procedure (4, 6, 8, 10 from Figure 1) show this iterative agreement method for data segments to be coded, codes to be used, and the placing of data segments into the same codes. Thus, our final code book was developed, taking into account the thirteen cases, not just a subsample of our data (see Patton 1990 and Seale 1999 for a discussion about reliability and validity in qualitative research).

276

FIELD METHODS

QUANTITATIVE TECHNIQUES TO SEARCH FOR CROSS-CASE PATTERNS Once we had the final code book, we focused our attention on analyzing founders’ values. We filtered the code family values for the thirteen cases and obtained a case-by-value frequency table, where the cells represent the number of times each value was mentioned by each founder. Table 1 shows the results. We analyzed this table using the nonmetric multidimensional scaling routine (MDS) in SPSS to explore visually the relationship between codes (values) and primary documents (family firm’s founders; see Kruskal and Wish 1978 on MDS. MDS requires a proximity matrix. We produced a dissimilarity matrix of chi-squared distance between objects (Dillon and Goldstein 1984:124). The chi-squared distance is similar to the Euclidean distance, with the distinction that each squared difference between profiles is weighted by the corresponding element of the average founder profile. di , i ′ =

∑∀

j

( f i , j − f i ′ , j )2 / f j ,

where fi,j is the i founder profile for the value j, that is, ni,j /n+j, and each squared difference between values is weighted by the corresponding element of the average founder profile, fj = n+j /n (Greenacre 1993:24–31). Afterward, we used Ward’s hierarchical clustering (Dillon and Goldstein 1984:172–78) to create a founders’ taxonomy (Step 13 in Figure 1). Although this technique was developed to analyze much larger data sets, it may be useful in exploring the structure embedded in a few cases. Finally, we returned to Atlas.ti and analyzed qualitatively the association between founders’ values and their family backgrounds. The nonmetric multidimensional solution (Kruskal and Wish 1978) obtained by using SPSS’s Alscal procedure produces a stress value of 0.265 for two dimensions (Young’s S-stress formula 1). Thus, a two-dimensional solution does not recover the structure in the original data. Adding more dimensions to the MDS did not significantly reduce the number of hard-to-fit items, but it did, of course, increase the complexity. Figure 3 plots the values in the two-dimensional solution. For ease of interpretation, values have been labeled with a number, with the meaning of each value number listed in figure note. The first dimension is associated with the values of business orientation, entrepreneurship, ambition, growth, and sacrifice in the negative direction and negative human relations, family orientation, cosmopolitan orientation,

TABLE 1 Contingency Table of Values and Founders Primary Document Value

277

Cosmopolitan orientation Altruism Ambition Autonomy Risk aversion Negative human relationships Positive human relationships Constancy External control Internal control Entrepreneurship Growth Dependence Determination Egotism Sense of family Stability Gratitude Honesty Innovation Economic interest Opportunistic orientation

1

2

3

4

5

6

7

8

9

10

11

12

13

Total

7 6 0 3 0 11 0 2 0 4 0 0 0 9 0 2 0 1 0 8 2 0

2 1 1 1 0 0 0 0 2 2 0 5 0 14 0 1 2 0 3 1 1 0

1 1 3 0 4 0 1 5 0 1 0 9 0 0 0 5 1 1 1 2 0 0

3 0 1 0 0 2 0 6 0 1 0 1 0 0 0 2 4 0 3 0 2 0

3 1 4 2 0 0 2 4 0 4 8 5 0 7 0 0 0 0 0 0 0 0

0 0 1 0 0 0 0 0 0 0 0 1 0 0 0 0 1 0 0 2 1 0

0 0 0 3 4 0 3 1 1 1 1 2 0 4 0 5 1 3 0 4 1 0

0 0 2 1 1 0 0 0 3 1 2 2 0 0 0 0 0 0 1 3 0 0

5 1 0 1 1 1 0 1 0 0 1 5 0 0 0 3 1 0 0 7 0 0

0 3 3 0 0 1 1 2 1 2 2 2 0 1 0 0 0 1 1 0 2 0

1 0 2 0 0 0 0 1 0 2 0 4 0 1 0 2 3 0 0 1 0 0

1 2 8 0 0 1 0 1 0 2 1 0 0 3 0 1 0 0 2 6 3 0

0 1 10 1 0 0 0 2 2 2 5 7 0 3 0 3 0 0 2 0 0 0

23 16 35 12 10 16 7 25 9 22 20 43 0 42 0 24 13 6 13 34 12 0

(continued)

278

TABLE 1 Continued Primary Document Value Short-term orientation Business orientation Ethical orientation Family orientation Local orientation Long-term orientation People orientation Task orientation Negative failure perception Positive failure perception Risk taking Rigor Sacrifice Satisfaction Simplicity Sense of guilt Sense of achievement Seriousness Hard work Active life Contemplative life Prosperous life Vocation Total

1

2

3

4

5

6

7

8

9

10

11

12

13

Total

0 0 1 8 0 2 0 1 0 0 0 0 0 3 0 2 10 2 8 0 0 1 0 93

0 0 4 3 0 0 2 0 1 4 0 6 0 4 0 0 4 2 5 0 0 0 0 71

0 3 4 2 0 2 0 0 0 0 0 1 3 1 0 0 1 0 2 0 0 0 0 64

0 0 1 10 0 0 0 0 0 1 0 1 1 2 0 0 0 3 5 0 0 1 0 50

0 3 0 1 0 4 0 0 0 0 0 0 1 1 0 0 3 0 2 0 0 0 0 55

0 3 0 2 0 0 0 0 0 0 1 0 2 0 0 0 0 1 5 0 0 0 0 20

1 2 1 9 0 2 0 0 0 1 0 0 2 6 3 0 0 2 3 3 0 3 1 73

0 4 0 0 0 1 1 0 0 0 0 0 2 1 1 1 2 2 2 1 0 0 0 34

0 1 1 3 0 1 0 0 0 0 2 0 1 3 2 0 1 0 1 1 0 1 0 45

0 5 5 0 0 1 3 0 0 2 0 0 1 1 0 0 0 1 3 2 0 0 0 46

0 5 1 2 1 1 0 0 0 1 1 1 2 3 0 0 1 0 1 1 0 0 3 41

0 1 0 4 1 1 2 0 0 2 0 2 2 1 0 0 1 5 8 4 0 3 2 70

0 8 2 2 1 2 3 0 0 2 0 1 3 3 2 0 3 3 10 2 0 0 0 85

1 35 20 46 3 17 11 1 1 13 4 12 20 39 8 3 26 21 55 14 0 9 6 747

García-Álvarez, López-Sintas / CONTINGENCY TABLE

279

FIGURE 3 Plot of the Founders’ Value Systems in Euclidean Space

NOTE: 1 = altruism; 2 = ambition; 3 = autonomy; 4 = risk aversion; 5 = negative human relationships; 6 = positive human relationships; 7 = constancy; 8 = external control; 9 = internal control; 10 = entrepreneurship; 11 = growth; 12 = determination; 13 = sense of family; 14 = stability; 15 = gratitude; 16 = honesty; 17 = innovation; 18 = economic interest; 19 = cosmopolitan interest; 20 = short-term orientation; 21 = business orientation; 22 = ethical orientation; 23 = family orientation; 24 = local orientation; 25 = long-term orientation; 26 = people orientation; 27 = task orientation; 28 = negative failure perception of failure; 29 = positive failure perception of failure; 30 = risk taking; 31 = rigor; 32 = sacrifice; 33 = satisfaction; 34 = simplicity; 35 = sense of guilt; 36 = sense of achievement; 37 = seriousness; 38 = hard work; 39 = active life; 40 = prosperous life; 41 = vocation.

task orientation, sense of guilt, economic interest, and innovation on the positive side. This first axis appears to reflect how founders see their firm: business as an end (the left orientation) and business as a means for the family. In our sample, if founders are located on the business-as-end orientation, the firm has a predominant role over the family. For those located on the business-as-means orientation, the aims of the firm are constrained by the family. The second axis is associated with risk aversion, constancy, sense of family, stability, and satisfaction in the negative direction and determination, sense of achievement, positive failure perception, people orientation, ambition, internal control, seriousness, and hard work in the positive direction. This dimension could be labeled group orientation (the southern direction) and self-fulfillment orientation (the northern direction). If founders are

280

FIELD METHODS

FIGURE 4 Single Plot of Founders in the Values’ Space (Dimensions 1 and 2)

NOTE: 1st Axis = Business Dimension (positive axis, business as a means; negative axis, business as an end). 2nd Axis = Psychosocial Dimension (positive axis, self-fulfillment values; negative axis, group orientation values).

located on the group orientation values, their actions are constrained by the desire to be accepted by the group, and family interests play an important role in their business decisions. On the other hand, those located on the selffulfillment orientation look for personal realization, and their family probably plays a minor role in their business decisions. To check methodological convergence (triangulation), we ran correspondence analysis on the founder-by-values matrix (Greenacre 1993). This method looks for an exact representation of the distances between values and founder profiles and plots cases and attributes in the same space. There is a clear coincidence in the MDS and correspondence solutions, although there is some discrepancy as well. The first two axes in correspondence analysis account for only 35.76% of the association between the group of founders and the group of values, whereas nonmetric MDS explains far more of the original order implied in the derived chi-squared distances. The relationship between the two procedures lends internal validity to our findings.

García-Álvarez, López-Sintas / CONTINGENCY TABLE

281

FIGURE 5 Clustering of Founders’ Profiles

NOTE: 1st Axis = Business Dimension (positive axis, business as a means; negative axis, business as an end). 2nd Axis = Psychosocial Dimension (positive axis, self-fulfillment values; negative axis, group orientation values).

When looking at the founders’ position in the two-dimensional MDS solution (stress is 0.195 for two dimensions, axes, and 0.091 for three), we found that founders are as far from the origin as their profile is from the founders’ average profile (see Figure 4). The first axis separates founders who view business as an end (5, 13, 10, 8, 11, 7, and 3) from those who view business as a means to their family life (2, 1, 12, 6, 9, and 4). The second axis splits the sample between founders who seek self-fulfillment values (cases 1, 2, 5, 12, 13, and 10) and those who seek social acceptance, group orientation values (cases 6, 8, 11, 9, 4, 3, and 7).

A TAXONOMY OF FOUNDERS From inspection of the plots, it appears that we can group founders according to their position in one of the four quadrants. Using Ward’s hierar-

282

FIELD METHODS

chical clustering procedure (Dillon and Goldstein 1984) with the original dissimilarities also suggests the formation of four groups of family firms’ founders (see Figure 5). One group comprises just a single founder. It may not appear logical for a cluster to have a single member. However, this is compatible with cluster analysis, and we maintain it because in-depth cross-case analysis shows that, in fact, Founder 1 is different from the members of the other three clusters. With so few cases, we may not have had sufficient data to achieve a taxonomy of four parts, as our theory suggested we would find. Further research, using the same methodology and larger samples, is needed to assess the usefulness of our taxonomy in other socioeconomic contexts. To label each group of founders, we took into account not only the previously mentioned values but also the specific characteristics of each cluster of founders. To search for these peculiarities, we went back to our original text data and performed a cross-case analysis of each group, taking into account all the codes from our code book. We looked for intragroup similarities, coupled with intergroup differences. We worked without labels for each group of founders until we completed the cross-case analysis for all the codes. Finally, we labeled the groups according their characteristics as founders of family tradition, founders-strategists, founders-achievers, and one founderinventor. The main characteristics of these groups are summarized below. Founders of Family Tradition Group values and the idea of the business as an end dominate the value system of these founders. They have a high frequency of values such as positive perception of human relations, ethical orientation, and family sense. Founders-Achievers The members of this group view their firms as a means to earn a family living and, as with founders of family tradition, the group-values orientation prevails in their value system. They have a high frequency of values such as short-term orientation, vocation, and task orientation. Founders-Strategists Although this group sees the business as an end, like founders of family tradition, self-fulfillment values predominate in their value system. They have a high frequency of values such as sense of achievement, internal control, and long-term orientation.

García-Álvarez, López-Sintas / CONTINGENCY TABLE

283

Founder-Inventor As with the case of founders-strategists, self-fulfillment values predominate in this value system, although the founder views his business as a means to earn a family living. Values such as innovation, family orientation, and a negative perception of human relations prevail.

QUALITATIVE EXTERNAL VALIDATION OF OUR CROSS-CASE PATTERNS: FAMILY BUSINESS BACKGROUND Once we settled on the labels for and value content of the groups of founders, we focused on the relationship between each set of founders and variables that defined their background as a way to identify the distinctive characteristics of our taxonomy. We next present textual evidence (verbatim quotes) that points out similarities and differences among the cases. The code family business background refers to the fact that founders have family ancestors with business experience. Most founders who saw their firms as an end (family tradition and strategist) mentioned relatives who were entrepreneurs; however, those who saw their firms as a means (achievers and inventor) had no previous family business experience. The existence of a previous family business background had a different meaning for the founder of family tradition and for the strategist. The former saw his business activity as something natural that had been part of his life since childhood: He had a small shop in the house and that’s where I was born, in this house and this shop. In those years, I had no other option and I gradually began to like the profession. It’s a vicious circle, the more you like something, the more you learn and the more you learn, the more you like it. (3:3 [15:23])

For the strategist type, this prior family business background also represents a challenge to reach more ambitious goals. If we study the presence of a family business activity background, namely, the existence of ancestors who worked in the same activity as the founders, this characteristic again differentiates founders of family tradition and strategists. Founders of family tradition saw their firms as a way to continue pursuing the activity of their families, and strategists saw the firms as a way to achieve self-fulfillment or seize an opportunity: . . . is a continuation of my father’s profession. (3:2 [14:15])

284

FIELD METHODS

The only thing I did was to expand the business, not only in terms of size but also in terms of geographic scope. (8:2 [20:23])

Achievers and the inventor without any family background began their businesses from scratch. In these cases, we identified a prevalence of negative factors in their start-ups, relating mainly to the impossibility of finding a job that matched their knowledge and skills: I encountered a lot of difficulty in finding work because what I know how to do isn’t done here. (1:130 [116:118]) Since there were no employment possibilities around here, I set up business. It wasn’t much of a business at the start, I did everything. . . . I fixed cars, boats, anything I could and gradually I got more involved until I began to construct boats. (1:124 [33:36])

We can see that the origin of the firm was completely different for each group of founders and clearly reflected their concept of the firm as an end or as a means. Achievers and the inventor set up their businesses as a way to earn a living for themselves and their families. We can also identify differences among groups of founders in our sample according to their education, which ranged from a complete lack of formal studies to university degrees. Founders of family tradition and strategists had a higher level of formal education than did achievers and the inventor. This can be linked to their family background, which provided them with business experience and financial resources to acquire an education. However, founders agree that their main source of business knowledge stems from their own experience: When I was 16, I went to work in Orense and worked at the plant, in the shop of my brother-in-law. I took advantage of my free-time in the evenings to get some training . . . in business math, to prepare myself for business a little. (2:21 [198:203]) I think from my reading. At the level of abstract ideas. In terms of the day-to-day, I think it was from my father, from my family in terms of living in a business environment, from other persons with whom you work throughout your life, as you rub shoulders with competitors, suppliers, customers. You have to learn from all these individuals, who apparently know more than you about some things. From living. (8:13 [165:173]) From a business point of view, nothing. In terms of academic training, I had studied fine arts. I had learned a little about design work because I had worked in a French company. I had reached an agreement in which they supported me in the entire manufacturing process and business management with their name and their product, and that’s where I started. But in terms of business training,

García-Álvarez, López-Sintas / CONTINGENCY TABLE

285

zero. Lots of will and desire to . . . believe in what I could develop and lots of desire to surround myself with people who would fill in what I was missing. (5:5 [77:92])

Although the founders mentioned that learning from experience was important for the development of their businesses, this doesn’t mean that they had previous business experience. Most founders in our sample didn’t own a previous business: They developed their business activity over the years based on their first start-up and didn’t learn from previous business failures, as suggested in some literature. Another aspect that the literature considers important for a successful entrepreneurial career is relying on the help and guidance of a mentor. However, only four of the founders in this sample mentioned them.

CONCLUSIONS This article shows one way to combine qualitative and quantitative instruments to obtain complementary findings about the lives of business founders. Qualitative data analysis software was very useful. It made data management and thematic coding easier and, by exporting a data matrix, it facilitated quantitative analysis as well. The creation of a taxonomy of founders helped us point out relationships in our data that would have been difficult to find using only qualitative techniques. Simultaneously, we were able to associate verbatim texts with the results of our quantitative analysis. This helped us perform in-depth cross-case analysis that further revealed underlying relationships. The next step in this iterative process will be to assess (confirm or disconfirm) the validity of our results by testing them on independent samples of data in Spain and in other cultural contexts.

REFERENCES Bernard, H. R., and G. W. Ryan. 1998. Qualitative and quantitative methods of text analysis. In Handbook of methods in cultural anthropology, edited by H. R. Bernard, 595–646. Walnut Creek, CA: Altamira. Cappuyns, K., M. A. Gallo, and A. Vilaseca. 1998. Successful family business. Paper presented at the Ninth Family Business Network Conference, Paris. Corbetta, G. 1999. Los valores de empresa y los negocios familiares. In Consideraciones éticas sobre la iniciativa emprendedora y la empresa familiar, coordinated by D. Melé Carné, 157–72. Pamplona, Spain: Ediciones Universidad de Navarra.

286

FIELD METHODS

Dillon, W. R., and M. Goldstein. 1984. Multivariate analysis—Methods and applications. New York: John Wiley. Doerfel, M. L., and G. A. Barnett. 1999. A semantic network analysis of the International Communication Association. Human Communication Research 25:589–603. Donckels, R., and E. Fröhlich. 1991. Are family businesses really different? European experiences from Stratos. Family Business Review 4 (2): 149–70. Dyer, W. G. 1986. Cultural change in family firms. Anticipating and managing business and family transitions. San Francisco: Jossey-Bass. Gallo, M. A., and C. García Pont. 1989. La empresa familiar en la economía española. Papeles de Economía Española 39/40:67–85. Glasser, B. G., and A. L. Strauss. 1967. The discovery of grounded theory. Strategies for qualitative research. New York: Aldine de Gruyter. Gómez, J. L., and J. C. Martínez. 1992. Las primeras 500 empresas de Galicia. La Coruña, Spain: Biblioteca Gallega. Greenacre, M. J. 1993. Correspondence analysis in practice. San Diego, CA: Academic Press. Jang, H., and G. A. Barnett. 1995. Cultural differences in organizational communication: A semantic network analysis. Bulletin de Methodolgie Sociologique 44:31–59. Jehn, K. A, and L. Doucet. 1996. Developing categories from interview data: Text analysis and multidimensional scaling. Part 1. Cultural Anthropology Methods Journal 9 (1): 15–16. ———. 1997. Developing categories for interview data: Consequences of different coding and analysis strategies in understanding text: Part 2. Cultural Anthropology Methods Journal 9 (1): 1–7. Kruskal, J. B., and M. Wish. 1978. Multidimensional scaling. Sage University Paper series on Quantitative Application in the Social Sciences, pp. 7–11. Beverly Hills, CA: Sage. La Fuente, F. A., F. V. Salas, and S. R. Pérez. 1985. Tipos de empresario y de empresa. El caso de las nuevas empresas españolas. Economía Industrial November/December:139–51. Monge, P. P., and E. M. Eisenberg. 1987. Emergent communication networks. In Handbook of organizational and management communication, edited by F. Jablin, L. Putnam, K. Roberts, and L. Porter, 204–42. Beverly Hills, CA: Sage. Muhr, T. 1997. ATLAS/ti, User’s manual and reference, Version 4.1. Berlin, Germany: Scientific Software Development. Patton, M. Q. 1990. Qualitative evaluation and research methods. London: Sage. Rice, R. E., and J. A. Danowski. 1993. Is it really just like a fancy answering machine? Comparing semantic networks of different types of voice mail users. Journal of Business Communication 30:369–97. Seale, C. 1999. The quality of qualitative research. London: Sage. Stohl, C. 1993. European managers’ interpretation of participation: A semantic network analysis. Human Communication Research 20:97–117. Strauss, A. L., and J. Corbin. 1998. Basics of qualitative research: Techniques and procedures for developing grounded theory. 2d ed. Thousand Oaks, CA: Sage. Ward, J. L. 1987. Keeping the family business healthy: How to plan for continuing growth, profitability, and family leadership. San Francisco: Jossey-Bass. ———. 1999. Valores que sintetizan las empresas familiares. In Consideraciones éticas sobre la iniciativa emprendedora y la empresa familiar, coordinated by D. Melé Carné, 149–55. Pamplona, Spain: Ediciones Universidad de Navarra.

ERCILIA GARCÍA-ÁLVAREZ, Ph.D., is a post-doctoral research fellow at the Family Business Centre at IESE business school. Her research is funded by the Secretaría de

García-Álvarez, López-Sintas / CONTINGENCY TABLE

287

Estado de Educación of the Spanish government and by the European Social Fund. Her research interests include computer-aided qualitative analysis, family business, and marketing of cultural products. JORDI LÓPEZ-SINTAS is a full professor in marketing and management at the Universitat Autònoma de Barcelona (UAB). His research interests concern the marketing of cultural products, small business economics, the use of quantitative models in marketing decisions, and computer-aided qualitative analysis. The last three publications Drs. García-Álvarez and López-Sintas have coauthored are “A Taxonomy of Founders Based on Values: The Root of Family Business Heterogeneity” (Family Business Review, 2001), “The Consumption of Cultural Products: An Analysis of the Spanish Social Space” (Journal of Cultural Economics, 2002), and “Omnivores Show Up Again: The Segmentation of Cultural Consumers in the Spanish Social Space” (European Sociological Review, 2002).