Ismo Koponen, Mr.
M.Sc. (Econ. - Intl. Marketing). Freelance Researcher -
Doctoral Candidate at
Lappeenranta University of Technology, Finland
Lecturer of Marketing and International Trade at
Oulu Polytechnic, Finland
Home: Teuvo Pakkalan katu 2 as 4, FIN-90100 Oulu
GSM: +358-(0)-500-929066
MEASURING CULTURAL DISTANCES;
FURTHER DEVELOPING AND TESTING
THE DYNAMIC METHOD –
A Swedish-Finnish-Russian Case Study
Report to be presented to the
Doctoral Tutorial related to the 7th Vaasa Conference on International Business at the University of Vaasa August 24, 2003
Abstract. The general purpose of my research project is to quantify The AIKA Component, a human factor of four determinants that seem to affect international cultural distances either narrowing or widening gaps between business organizations. The determinants of the component are: Attitude, Interest, Knowledge and Attitude Adaptation. My other assumptions include e.g. the idea that the above determinants are all interrelated. For procedures of quantitative data with interrelated determinants I have developed The Koponen Manipulator, a multidimensional scaling instrument. It should be of some use with other mathematical applications, including Multi Criteria Decision Making (MCDM).
The more specific aim of this very paper is to report on my study on the Mini Triad of Sweden, Finland and Russia. A survey (n=65) was made in Finland - the source business culture - to test the dynamic method. The target business cultures are Sweden and Russia. The outcome of the study supports the idea of dyadic cultural distances.
Key words: AIKA Component, cultural distances, cultural properties, Dynamic Method, international business, Koponen Manipulator and partner selection.
11
Acknowledgements: I wish to thank the Foundation for Economic Education (Liikesivistysrahasto) and the Marcus Wallenberg Foundation. They both support my project for a doctoral thesis. Mrs Aira Davidsson from FINTRA, Helsinki, Finland, has organized the collecting of the data. Thank you.
Introduction: This is a cross-cultural case study that I base on some empirical data collected in Finland. The very meaning of the study is to contribute to cultural distances model development and to quantify the psychic distance from Finland to Sweden and Russia.
The report is divided into three chapters. 1.) In the first one, I will briefly introduce the present paradigm of cultural differences and describe the three focal cultures with help of Geert Hofstede’s four cultural dimensions. In addition to this I will introduce my own AIKA factor with its four determinants. 2.) In the second chapter I will introduce The Koponen Manipulator and explain my other methodological choices. 3.) Chapter three is reserved for the data analyses. The two pages of appendices include a full SPSS component matrix and two box plots based on cross-tabulation. I refer to the appendices in my text.
1. Cross-Cultural Research in Present Theories
Because I am going to use Hofstede’s four dimensions for describing the cross-tabulation of Sweden, Finland, and Russia it is necessary to familiarize ourselves with some of his ideas. Here are a few statements. About doing cross-cultural research, in general. “...in studying ‘culture’ we compare societies (Hofstede, 15). “...studying differences in culture among human groups and categories that think, feel, and act differently does presuppose a position of cultural relativism” (ibid. 21). “Cultural relativism does not imply normlessness for oneself or for one’s own society. It does call for one to suspend judgement when dealing with groups or societies different from one’s own” (ibid.).
How distant are the societies or cultures in question, is a general level question that my research on the AIKA factor deals with. The determinants of the factor are Attitude, Interest, Knowledge, and Adaptation ability. It is a coincident that the word ‘aika’ in my native Finnish language has the meaning of ‘time’. Time is, indeed, an important dimension of any dynamic phenomenon; thus, also, of the cultural distances between international business organizations. ...attitudes toward a behavior are found to correlate well with the corresponding behavior, and since they can be assessed ahead of time they can be used to predict behavioral performance” (Aizen, 1988: 109).
11
But, back to Hofstede: “Language is both the vehicle of most of cross-cultural research and part of its object. The problems with the use of language in research about culture start before the actual translation of questions. The researchers and their informants may behold different normative expectations about the use of language” (2001: 21). The developers of the semantic differential (a research instrument described later) evaluate the ‘psycholinguistic’ instrument: “...we do not find any appreciable difference between factors in terms of reliability. Cultural meanings of concepts prove to be very stable - for any factor, a shift of only about four tenths of a scale unit is significant for at the 95 per cent level. This degree holds despite the small sizes of the groups, only about 25 in each” (Osgood et al, 1957: 139-140). Their scale had 7 units. Later, Osgood with other co-authors, ask: “Can we avoid the problem of subjective judgment based on translation?” (1975: 31). Their answer is: “Subjects are not the same, ... and scales are obviously not the same, ... but culture-common concepts are at least closely translation-equivalent” (ibid). I will proceed with a certain feeling of confidence.
Hofstede’s own research covers all the three cultures of the mini triad in focus. Data (Index Score Estimates) on Russia were added in 1996 (see e.g. 2001: 502).
Table 1. Culture Index Scores of Hofstede’s Dimensions
Dimension Culture / Power Distance / Uncertainty Avoidance / Individualism / Collectivism / Masculinity / FemininityAustria / 11 / 90
Denmark / 23
Finland / 33 / 59 / 63 / 26
Greece / 112
Portugal / 27
Russia / 93 / 95 / 39 / 36
Slovakia / 104 / 110
Sweden / 31 / 29 / 71 / 5
In the above table there are all the scores for the cultures of the mini triad and some European extremes, dimension by dimension.
11
Sweden hits one record with her low masculinity index, other record scores going to cultures outside the triad in focus. Row sums of the scores for the triad cultures are: Sweden 136, Finland 181, and Russia 263. The score difference between Sweden and Finland is 45 points. It is 82 points between Finland and Russia, and 127 between Russia and Sweden.
This can give an idea about cultural distances between the triad’s nations. One should, however, be careful with the possible prediction that all the cultural dimensions be on the same vector.
2. Developing and Exploiting the Research Instrument
Before going to the measurements of attitude and other psychic determinants, it is necessary to review a few ideas concerning attitude measurement. Ajzen argues e.g., that “An attitude is a disposition to respond favorably or unfavorably to an object, person, institution, or event”, and that:
“Like personality trait, attitude is hypothetical construct that, being inaccessible to direct observation, must be inferred from measurable responses. Given the nature of the construct, these responses must reflect positive or negative evaluations of the attitude object. Beyond this requirement, however, there is virtually no limitation on the kinds of responses that can be considered” (1988: 4).
An attitude is obviously based on a person’s beliefs, or as Fishbein and Ajzen put it: “...a persons’ attitude toward some object is determined by his beliefs that the object has certain attributes and by his evaluations of those attributes” (1975: 14). “The concept ‘attitude’ should be employed only where there is clear evidence that the obtained measure places the concept on a bipolar affective dimension” (ibid, 56).
Some scale progression. As a measuring scale for every survey question I have used a so called visual analogue scale (VAS) consisting of 21 ‘pixels’. With a verbal expression in both ends of the scales they have formed a variation of the semantic differential scale e.g. like this:
abcde (------) fghij
The positive poles were always at the left end of each scale and the negative ones at the right end of them. This is typical for the semantic differential scales, and I have been ‘obedient’. On my opinion the other way round might be a more logical or a more mathematical way of doing this. Nevertheless, the 21 pixels of the scales make it possible to consider every scale a true bipolar scale with a zero in the center:
+2, +1, 0, -1, -2.
11
Since it is important to determine the exact position of zero I have reserved one (whole) pixel for it. If we judge the technique mathematically, this is wrong because zero, actually, has no dimensions at all; zero can not be measured. This is, of course, a problem with any bipolar scale that reserves a space for zero - usually as wide as for any true digit.
The VAS has made it possible for me to consider the scale a four point unipolar one, ranging between four and one ( 4, 3, 2, 1). This was important for me because I wanted to avoid ‘messing’ with zero, this time. One benefit of this VAS is that I may, later, rescale the instrument - without any violation to data - returning to the intended bipolar scale. Unipolar or bipolar, both the above numeric expressions rest on a semantically bipolar scale. As stated before, bipolar scales are a necessity e.g. by Fishbein and Ajzen. They argue that “Our definition of attitude requires a measurement procedure whereby a person assigns some concept to a position on a bipolar evaluative dimension” (1975: 56). A few markings on the very centre of my VASes confused me first but then I decided to read these markings two-point-five.
The Koponen Manipulator. The following table illustrates the data procedures of the manipulator. In the table, columns A, B, C and D stand for four determinants, column RS for row sums. The surveyed frequency values lay on row one (1). Equivalent weights for the determinants are on row two (2). The determinants can be e.g. surveyed or given. Row three (3) holds the results of R1 x R2 multiplications. All excess weight (49.30 - 14.39 = 34,92) is evenly (34,92 / 4 = 8,73) destructed from the previous multiplication: R3 - R4 = R5. Row five holds the final - manipulated - averages.
Table 2. Experimenting with The Koponen Manipulator
A / B / C / D / RSR1 = / 3.64
(1.) / 3.51
(4.) / 3.61
(3.) / 3.63
(2.) / 14.39
R2 x / 3.75 / 3.39 / 3.14 / 3.42
R3 = / 13.65 / 11.90 / 11.34 / 12.41 / 49.30
R4 - / 8.73 / 8.73 / 8.73 / 8.73
R5 = / 4.92
(1.) / 3.17
(3.) / 2.61
(4.) / 3.68
(2.) / 14.38
11
A few important remarks. As seen from the above table, the manipulation procedure does not affect the imaginary ABCD factor’s total value (14,39 or 14,38) but the individual determinant values have changed. The factor is a ‘closed context’ where all determinants affect all determinants i.e. they are interrelated. If the example described an MCDM case, determinant A might indicate a strength and e.g. C a weakness.
3. The Survey and Analysing the Data
About the sample. The 65 answerers of the survey represent firstly the clientele of FINTRA, i.e. Finnish business people that have lived true the process of belief, attitude, intention, and behavior concerning education in international business at the particular institution, and secondly - more generally - business people from around the nation.
Table 3. The Sample Quantified by Gender and Generation, (n=65)
Demographics / 1. Gender: Male (xx%) / 2. Gender: Female (xx%)1. Generation:
X = Junior (xx%)
- age: up to 30 years / xx / xx
2. Generation:
Senior (xx%)
- age: 31 years + / xx / xx
The juniors represent the so called Generation X, that are highly computerized and communicate wirelessly. The youngest respondent was xx years of age, and the oldest xx. Twenty of the 65 answerers held a managerial position in his or her organization, 41 were in a non-managerial position, and 4 did not want to answer to the question.
The parts of the AIKA component are all positive. Together they indicate a self attributed resource or force the respondents feel they have for fighting the cultural distance to another nation.
11
Table 4. The AIKA Factor on Psychic Distance From Finland to Sweden (n=65)
Attitude / Interest / Knowledge / Adaptation Ability / RSAverage / 3.13 (1.) / 2.66 (4.) / 2.78 (3.) / 3.10 (2.) / 11.67
Weight X / 3.75 / 3.39 / 3.14 / 3.42
= / 11.74 / 9.02 / 8.73 / 10.62 / 40.09
- / 7.11 / 7.11 / 7.11 / 7.11
Manipulated / 4.63 (1.) / 1.91 (3.) / 1.62 (4.) / 3.49 (2.) / 11.67
The Finnish respondents think that the cultural distance from Finland to Sweden is 2,12 (max 4,00). This figure can be reflected with any single determinant of the above table (R1: Average, or R5: Manipulated). An appropriate parallel figure for the component’s total value (11,67) could be gained by multiplying 2,12 by 4 (=8.48). The positive force of 11,67 covers well the distance of 8,48. Please, pay attention to the fact that the data manipulation has changed the order of total value of the determinants.
Table 5. The AIKA Factor on Psychic Distance From Finland to Russia (n=65)
Attitude / Interest / Knowledge / Adaptation Ability / RSAverage / 2.48 (2.) / 2.53 (1.) / 2.31 (3.) / 2.16 (4.) / 9.48
Weight X / 3.75 / 3.39 / 3.14 / 3.42
= / 9.30 / 8.58 / 7.25 / 7.39 / 32.52
- / 5.76 / 5.76 / 5.76 / 5.76
Manipulated / 3.54 (1.) / 2.82 (2.) / 1.49 (4.) / 1.63 (3.) / 9.48
The Finnish respondents think that the cultural distance from Finland to Russia is 2,91 (max 4,00). This figure can be reflected with any single determinant of the above table (R1: Average, or R5: Manipulated). An appropriate parallel figure for the component’s total value (9,48) could be gained by multiplying 2,91 by 4 (=11,64). The positive force of 9.48 does not fully cover the distance of 11,64. Please, pay attention to the fact that the data manipulation has changed the order of total value of the determinants.