This chapter provides the reader with information on how the study was conducted;
it will touch upon issues such as the research framework, the hypotheses tested and the procedure in which these hypotheses were researched. It will also provide the reader with the questionnaires used to develop the research, the information about the sample, the data collection methods and the process in which the data was validated and analyzed.
Research Framework
The following figure illustrates the relationships that will be analyzed throughout the study. The five dimensions of Hofstede play the role of moderators in the relationships between PU, PEOU and SN and behavioral intention to use a system.
Long Term/ Short Term Orientation
Subjective Norms Perceived Ease of Use
Perceived Usefulness
Femininity/
Masculinity Collectivism/
Individualism
Uncertainty Avoidance Power Distance
Behavioral Intention
Figure 3.1. Research framework
30
This research aims to first investigate the national culture characteristics of the different countries in our sample and then analyze what kind of moderating role they play in the relationship between the TAM.
Hypotheses
Hypothesis 1:
Hypothesis 2:
Hypothesis 3:
Hypothesis 4:
Perceived usefulness of social networks for staffing activities will have a significant influence on behavioral intention to use this system.
Perceived ease of use of social networks for staffing activities will have a significant influence on behavioral intention to use this system.
Subjective norms will have a significant influence on behavioral intention to use social networks for staffing activities in the organization.
The relationship between subjective norms and behavioral intention to use is moderated by the espoused national cultural value of individualism/collectivism, so that the relationship is stronger for individuals with higher collectivist values.
Hypothesis 5a: The relationship between perceived usefulness of the social networks for staffing activities and behavioral intention to use is moderated by the espoused cultural value of uncertainty avoidance, such that the relationship is stronger for individuals with a lower espoused uncertainty avoidance.
Hypothesis 5b: The relationship between perceived ease of use of the social networks for staffing activities and behavioral intention to use is
31
moderated by the espoused cultural value of uncertainty avoidance, such that the relationship is stronger for individuals with higher espoused uncertainty avoidance cultural value.
Hypothesis 5c: The relationship between subjective norms and behavioral intention to use is moderated by the espoused national cultural value of uncertainty avoidance such that the relationship is stronger for individuals with higher espoused uncertainty avoidance cultural values.
Hypothesis 6: The relationship between subjective norms and behavioral intention to use is moderated by the espoused national cultural value of power distance such that the relationship is stronger for individuals with higher espoused power distance cultural values.
Hypothesis 7a: The relationship between perceived usefulness of the social networks for staffing activities and behavioral intention to use is moderated by the espoused national cultural value of masculinity/femininity such that the relationship is stronger for individuals with higher masculine cultural values.
Hypothesis 7b: The relationship between perceived ease of use of the social networks for staffing activities and behavioral intention is moderated by the espoused national cultural value of masculinity/femininity, so that the relationship is stronger for individuals with higher feminine cultural values.
Hypothesis 7c: The relationship between subjective norm and behavioral intention is moderated by the espoused national cultural value of masculinity/femininity, so that the relationship is stronger for
32
individuals with higher feminine cultural values.
Hypothesis 8a: The relationship between perceived usefulness of social networks for staffing activities and behavioral intention to use is moderated by the espoused national cultural value of long-term/short-term orientation, such that the relationship is stronger for individuals with espoused long-term orientation cultural values.
Hypothesis 8b: The relationship between subjective norm and behavioral intention to use is moderated by the espoused national cultural value of long-term/short-term orientation, such that the relationship is stronger for those individuals with espoused long-term orientation cultural values.
Research Procedure
Figure 3.2 presents the research procedure that will be followed during the development of this research. The literature review is one of the most important processes because this way the author was able to establish groundwork for the study.
After establishing the groundwork the decisions about the sample and the sampling procedure were made. HR practitioners are the target of this study because they have never been researched on the acceptance of a new technology.
After making the basic decisions about the study further investigation on the instrument to be used was made. Since this is a cross-cultural study the instrument was translated to other two languages. Expert review was requested, to make sure that the original meaning of the English version was kept in both the Spanish and Chinese version.
33
Afterward, the instrument was tested, we proceeded to the collection of the data;
data analysis was conducted after, and the findings of the study presented. The author also provides recommendations and suggestions for future research.
Research Design
This is an empirical study on Davis’ TAM; the original relationships presented in the TAM were tested, i.e. PU, PEOU, SN, and their relationships with behavioral intention to use the system. Besides, the moderating effect of culture using Hofstede’s cultural dimensions was tested in the relationship of PU, PEOU, SN and behavioral intention to use.
Human Resource practitioners were asked to complete a questionnaire in which both the technology and cultural constructs were included and tested. Quantitative statistical analysis was applied to test hypotheses.
Present findings, conclusions and recommendations for future research Data analysys
Data collection
Expert review and pilot test the instrument, modify when necessary Translate the Instrument to the different languages to be used
Find and adapt the instrument to the study
Make decisions about the sample and the sampling procedure Establish groundwork for the study
Review Literature
Figure 3.2 Research procedure
34
Sample Setting
The population researched in this study is HR practitioners of four different countries; these countries were chosen because of their different scores in Hofstede’s five dimensions of culture. Even though espoused culture will still be measured, it is important to show that there is already a precedent that has been seen on the differences of these countries. Table 3.1 presents Hofstede’s original scores for these four countries.
Four countries were chosen to ensure sufficient variance in their national cultural values. The intention of this research was not a country to country comparison, but just to examine the effect of culture in the acceptance of a technology.
Table 3.1.
Hofstede’s five dimensions of culture and its given scores
Country measure the constructs of the study. This helps the study to ensure higher reliability and validity.
TAM
Please refer to Table 3.2 to see the items used in the measurement of the TAM.
Perceived usefulness is defined as “the degree to which a person believes that using a particular system would enhance his or her job performance” p. 26; it is measured using 6 items developed by Davis (1989).
35
Perceived ease of use is defined as “the degree to which a person believes that using a particular system would be free of effort” p.26; it is measured using 6 items developed by Davis (1989).
Behavioral intention based on the Theory or Reasoned Behavior which states that intentions lead to action (Fishbein & Ajzen, (1975); measured using 2 items adapted from Srite and Karahanna (2006), BI1 and BI2.
Subjective Norm is defined as “a person’s perception that most people who are important to him think he should or should not perform the behavior in question”
(Fishbein & Ajzen, 1975, p. 302); it is measured using 3 items adapted from Srite and Karahanna (2006).
2. Using social networks would improve my performance in staffing the organization.
3. Using social networks in my job would increase my productivity for staffing the organization.
4. Using social networks would enhance my effectiveness on the job when staffing the organization.
5. Using social networks would make the staffing activities easier.
6. I would find social networks useful in my job.
Perceived Ease of Use
1. Learning to operate social networks for staffing purposes would be easy for me.
2. I would find it easy to get the social networks to do what I want them to do.
3. My interaction with social networks would be clear and understandable.
4. I would find social networks to be flexible to interact with.
5. It would be easy for me to become skillful at using social networks for staffing the organization.
6. I would find social networks easy to use.
Behavioral Intention 1. I intend to use social networks for staffing purposes.
2. I intend to use social networks frequently for staffing purposes.
(continued)
36
Subjective Norms
1. My coworkers think I should use social networks for staffing purposes.
2. Other HR practitioners think I should use social networks for staffing purposes.
3. My supervisor thinks I should use social networks for staffing purposes.
Culture
The cultural construct is based on Hofstede’s 5 dimensions:
individualism/collectivism, uncertainty avoidance, power distance, masculinity/femininity and long/short-term orientation. However, since the instrument used by Hofstede in his study was not directly created for the cultural analysis but to analyze job satisfaction in the IBM offices around the world, the new instruments developed later by Dorfman and Howell (1988), and adapted by Wu (2006) and Srite and Karahanna (2006) were used. All these three instruments were compared, then identical items were kept only once and those unique from each instrument were included.
Expert review on the content validity showed that the items used in the Masculinity/Femininity dimension were not appropriate to measure the espoused masculinity femininity dimension, because the items in these questionnaires referred to the genetic sense of masculine and feminine and not to the psychological sense of the words. These items included words as man and woman, misleading the respondents to a gender orientation instead of cultural values. Thus, the items were modified and additional items were created based on literature.
The dimension of power distance is measured using 10 items; the individualism/
collectivism dimension is measured using 7 items; the uncertainty avoidance dimension is measured using 6 items; the masculinity/femininity dimension is measured using 7 items and the long-orientation/short-orientation dimension is measured using 4 items.
Table 3.2. (continued)
37
Items have been taken from three different authors, so refer to table 3.4 for detailed information.
Agreement with the items of Individualism/Collectivism means high collectivism, since all the items are written to show the collectivism side of this dimension.
Agreement on the Uncertainty Avoidance items means high levels of uncertainty avoidance. On the Masculinity/Femininity dimension agreement with the items means higher masculinity levels. On the power distance dimension, agreement means high power distance levels. Refer to Table 3.3 to see the items in this dimension.
Table 3.3.
Culture Questionnaire Scales
Power Distance Author 1. Managers should make most decisions without consulting
subordinates.
(Dorfman & Howell, 1988)
2. It is frequently necessary for a manager to use authority and power when dealing with subordinates.
(Dorfman & Howell, 6. Managers should not ask subordinates for advice, because
they might appear less powerful.
(Srite & Karahanna, 2006) 7. Decision making power should stay with top management
in the organization and not be delegated to lower level privileges than lower level managers and professional staff.
(Srite & Karahanna, 2006) 10. A manager should perform work which is difficult and
important and delegate tasks which are repetitive and mundane to subordinates.
(Srite & Karahanna, 2006)
Individualism/Collectivism Author
1. Group welfare is more important than individual rewards. (Dorfman & Howell, 1988)
2. Group success is more important than individual success. (Dorfman & Howell, 1988)
(continued)
38
4.Employees should only pursue their goals after considering the welfare of the group.
(Dorfman & Howell,
6. Individuals may be expected to give up their goals in order to benefit group success.
(Dorfman & Howell, 1988)
7. Being accepted as a member of a group is more important than having autonomy and independence.
(Srite & Karahanna, 2006)
Uncertainty Avoidance Author
1. It is important to have job requirements and instructions spelled out in detail so that employees always know what they are expected to do.
(Dorfman & Howell, 1988)
2. Managers expect employees to closely follow instructions and procedures.
(Dorfman & Howell, 1988)
3. Rules and regulations are important because they inform employees what the organization expects of them.
(Dorfman & Howell, 1988)
4. Standard operating procedures are helpful to employees on the job.
(Dorfman & Howell, 1988)
5. It is better to have a bad situation that you know about, than to have an uncertain situation which might be better.
(Srite & Karahanna, 2006) 6. People should avoid making changes because things could
get worse.
(Srite & Karahanna, 2006)
Masculinity/Femininity Author
1. Being effective is more important than being social within the workplace. (Modified)
(Dorfman & Howell, 1988)
2. Achieving success and power is more important than receiving people’s approval or reaching consensus.
(Modified)
4. Solving organizational problems usually requires an active and forcible approach. (Modified)
(Dorfman & Howell, 1988)
5. My culture values recognition and promotion. (Modified) (Srite & Karahanna, 2006) 6. I would rather have a harmonious working environment
with a lower pay than a tense one with higher payment.
(Srite & Karahanna, 2006) 7. When trying to reach a goal being assertive is more
effective than being modest and caring.
(Srite & Karahanna, 2006)
Long/Short-Term Orientation Author
1. Ordering relationships by status and observing this order is important in the workplace.
(Wu, 2006)
2. Thrift is important in the workplace. (Wu, 2006)
3. Persistence is important in the workplace. (Wu, 2006) 4. Having a sense of shame is important in the workplace. (Wu, 2006) Table 3.3 (continued)
39
Demographics
Participants were asked to provide their nationality, age, gender, education level, industry for which they worked, whether they had a personal profile in a social network and also whether the company allowed the use of social networks in the workplace.
Control Variables
It is suspected that the organizational social network use may have an effect on the dependent variables; therefore this research included it in the analysis as a control variable.
Data Collection
Data was collected through online distribution of survey questionnaires; these questionnaires were created in a specialized website for questionnaires, and then a different URL was distributed for each country to assure that each sample was distinguished clearly.
The questionnaires were translated into 3 languages, Chinese to fit the Taiwanese sample, English for the Indian sample and Spanish for the Spanish and Guatemalan sample. The pilot test was conducted on 45 HR practitioners in different organizations both in Taiwan and Guatemala. This pilot was only used to ensure that the factor structure of the constructs wouldn’t change, since it didn’t the questionnaire remained the same and more sample was collected.
A seven-point likert-scale was used for measurement of agreement of the respondents where 1 represents strongly disagree and 7 represents strongly agree. The HR practitioners’ contact information in different countries was gathered by visiting the country websites of different companies and collecting the information of the HR
40
personnel. Later these personnel was contacted and asked to fill out the survey questionnaire. Social networks such as Facebook and LinkedIn were also used to contact the HR practitioners, through HR professionals’ groups.
Sample Profile
In this research questionnaires were distributed among four different countries, a total of 101 questionnaires were collected; 19% of the respondents were Indian, 24%
Taiwanese, 40% Guatemalan and 17% Spanish.
Seventy five percent of the respondents were between 20 and 39 years of age, and only 20% were above 40 years old but below 60 years old. Fifty-six percent of respondents were female and forty-four male.
All of the respondents had a college degree or above, this shows the importance of the education background in the Human Resource profession, none of the respondents reported to have education lower than college.
There were 94% of the respondents that reported ownership of a social network profile. There is still a lot of variation on the organizational social networks use (OSNUSE), 30% of respondents reported that their companies prohibit the use of social networks and 9% reported that their company’s demand the use of social networks for staffing activities.
As for the industries in which the respondents served, 28% of respondents belonged to the service industry, 16% to financial services and 15% to products manufacturing. The rest of the percentage was mixed among many industries.
Table 3.4 lists detailed information on the descriptive statistics for this study:
41 Table 3.4.
Descriptive Statistics of the Sample (N=101)
Variable Category Count Percentage
(%)
Education College degree or certificate 8 7.9
Bachelor degree 43 42.6
I rarely check my profile (less than
once a month) 5 5
I seldom check it (once a month) 9 8.9 I check it frequently (once a week) 24 23.8 I always check it (every day) 52 51.5
Organizational Size Up to 10 employees 9 8.9
Up to 50 employees 12 11.9
The company has strict practices that prohibit the use of social networks
30 29.7
The company has loose policies that prohibit the use of social networks
23 22.8
The company encourages the use
of social networks 10 9.9
The company owns a social
network profile 22 21.8
The company demands the use of social networks for staffing activities
9
8.9 (continued)
42 Table 3.4. (continued)
Variable Category Frequency Percentage
(%)
Industry Financial Services 16 15.8
Automotive Industry 2 2
Construction Industry 2 2
IT/Technology Industry 7 6.9
Products Manufacturing 15 14.9
Service Industry 22 27.8
Telecommunications 8 7.9
Consulting 6 5.9
Medical Industry 3 3
Other 18 17.8
Data Analysis Procedure
After the correlation analysis was conducted, the hypothesis testing took place.
PLS was used because it is a structural equation modeling (SEM) technique.
“Compared to regression-based approaches, which analyze only one layer of linkages between independent and dependent variables at the same time, SEM, as a second-generation technique, allows the simultaneous modeling of relationships among multiple independent and dependent constructs” (Haenlein & Kaplan, 2004, p. 285).
Haenlein and Kaplan (2004) state that regression-based approaches (e.g., multiple regression analysis, analysis of variance) and factor or cluster analysis, have limitations;
the limitation we are more concerned about is that these tools don’t allow us to analyze more complex and more realistic situations “this assumption of regression-based approaches may be too limiting for an analysis of more complex and more realistic situations. This becomes, for example, especially obvious when one wants to investigate the potential effect of mediating or moderating variables in the relationship between one or more dependent and independent variables, which may result in some dependent variables influencing other dependent variables” (Hainlein & Kaplan, 2004, p. 284).
43
This in our case, posed a serious concern in the analysis of the data, because the purpose of this research is mainly to investigate the moderating effect of culture in the acceptance of social networks for the staffing activities of an organization, thus PLS was used.
Another limitation of regression-based approaches is that each observation could contain measurement error, both from random error or systematic error. This could cause the scores to be formed by the sum of three parts, the true score, the random error and the systematic error (Churchill, 1979); first-generation techniques cannot be used when either of these errors occur, which is something that doesn’t happen often in reality.
PLS also has the advantage of allowing us to use both reflective and formative indicators, so that we can analyze those cases in which the indicators depend on the latent variable and those that cause the latent variable (Hainlein & Kaplan, 2004).
PLS is also recommended when the sample size is small, without a tool such as PLS the advisable sample size for SEM would be of about a 100 (Nasser & Wisenbaker, 2003), or at least 200 according to other researchers’ recommendations (Marsh, Balla &
Grayson, 1998).
Validity and Reliability
Even though previously validated scales were used in this study, tests were conducted to assure that the instrument effectively measured what it was intended to
Even though previously validated scales were used in this study, tests were conducted to assure that the instrument effectively measured what it was intended to