This chapter provides the image of the research framework of the assumption for the four constructs, which are the career adaptability, risk aversion, entrepreneurial opportunity, and entrepreneurial intention. The content is mainly describing the instruments that were selected to analyze the hypothesis, participants for analyzing the assumption and the reason to collect data from the targeted participants. After the data collection, the approaches for analyzing the data are listed in this chapter to reveal the relation of the constructs and the phenomenon of the selected participants.
Research Framework
Along with the definition of each constructs in the literature review, the research framework discloses the potential antecedents that could predict the entrepreneurial intention of selected participants. As mentioned in the hypothesis, the predictor is career adaptability, which is assumed to be predicting the criterion, entrepreneurial intention. With the moderating effect of risk aversion, a weakened relationship of the predictor and the criterion is proposed.
On the other hand, considering the moderating effect of the entrepreneurial opportunity, a strengthened relationship of predictor and criterion is hypothesized. The research framework is shown as Figure 3.1.
Figure 3.1. Research framework
Research Hypothesis
Base on chapter of literature review, which indicates that relationship between variables, the researcher proposes the following Hypotheses:
Hypotheses 1: Career adaptability is positively related to entrepreneurial intention.
Hypotheses 2: Risk aversion will weaken the relationship of career adaptability and entrepreneurial intention.
Hypotheses 3: Entrepreneurial opportunity will strengthen the relationship of career adaptability and entrepreneurial intention.
Career Adaptability (X)
Entrepreneurial Intention (Y)
Risk Aversion
(Mo
1)
Control Variable Age
Gender H1+
H2 -
Entrepreneurial Opportunity
(Mo
2)
H3+
Research Procedure
In order to investigate the study of the stated Hypotheses comprehensively, the study followed the eight-step research procedure, as shown in Figure 3.2. First, the researcher reviewed the literature for the specific topics. After the literature of the related topic is reviewed, the next step is to identify the research gap and the research problem. Then, the research purposes and questions were developed. Once the questions were stated, the research framework was established for the study, with the instrument based on mature measurements found in the literature review. Before the data collection, the pilot test was conducted to ensure the instrument was valid and reliable. Based on the analyzed data, the last procedure was to provide the results and findings in the study.
Figure 3.2. Research procedure
Review Literature Identify Research Topic
Develop Research Purpose and Questions Develop Research Framework
Develop Research Instrument Conduct Expert Review and Pilot Test
Collect and Analyze Data
Conclude Research Findings
Research Design
The research is giving insight to the entrepreneurial intention of Taiwan’s workers in different career status, including part-time and seasonal workers, contractors, temporary workers, full-time workers and dual-career professional workers. Thus, the Hypotheses of this study stated that career adaptability is the independent variable predicting the dependent variable, entrepreneurial intention; while the researcher also considered that risk aversion causes a weakening effect and entrepreneur opportunity a strengthening effect to the relationship between the independent variable and the dependent variable, according to the previous literature and theories. In order to test these Hypotheses, the study applied the quantitative method, by administering the questionnaires as the approach to collect the data.
As to ensure the validity and reliability of the measurements of the questionnaire, the study conducted a pilot test before massive data collection.
Sampling and Data Collection
The data collection was conducted from September till the end of November of 2018. The channel for collecting was web-based via social media. Several researchers studied the entrepreneurial intention antecedent and most research collect the data from undergraduate students or master students who participated in an entrepreneurship or leadership course, with the assumption that the individuals who have the knowledge of entrepreneurship tend to evoke a startup idea or self-employ concept. This study is probably the first to investigate various types of career status of the workforce, unlike the previous studies that used students as the sample. The study widely collected from workers in Taiwan using a convenience sampling approach, including part-time and seasonal workers, contractors, temporary workers, full-time
workers and dual-career professional workers. A question is needed to screen out the participants who were either an entrepreneur or a freelancer, by asking what their current career status was. The reason of screening out the entrepreneur and the freelancers was because that their responses to the questionnaire may cause a distortion on the scale of the entrepreneurial intention asking how strong the intention to be an entrepreneur, since they were certainly high on their intention to be an entrepreneur, when in fact they are.
The data collection was done through the method of using online survey, which was distributed through social media, such as personal Facebook and personal Line. In order to avoid the participant that was not qualified to fill out the questionnaire, the requirements were set as following: Taiwanese; currently not a student, and currently working or had working experiences. In addition, the questionnaire is distributed on the fan page for freelancers or start-ups on Facebook, besides posting on the page with the three requirements as stated above, the participants could not be an entrepreneur. The research also used the snowball approach to collect more data by requesting those in the personal network to approach the individuals who were working as a temporary worker, contractor, or part-time worker who met the requirements.
In the end, a total of 320 participated in the study, however, there were 2 duplicate responses, in addition to a total of 23 participants who were screened out for being an entrepreneur or a freelancer. Therefore, only 295 responses for the study were valid for inclusion in the data analysis.
Questionnaire Design
The questionnaire survey was designed with well-developed scale. The instruments were initially designed in English and published in Western journals. However, the participants of
the study are Taiwanese, whose official language is Mandarin; for this reason, the questionnaire was translated into Mandarin, to ensure the participants were able to understand each item properly. Since the questionnaire had to be translated into Mandarin, to remain the initial meaning of each items, the study used a translator who was fluent in English, with a TOEIC score above 860. After the translation, the questionnaire was reviewed by experts and backward translated by a peer who was proficient in both Mandarin and English, in order to ensure that the translation is consistent to English meaning. Moreover, the data collection used self-reported method, which has the possibility of Common Method Variance (CMV) that might affect the result of the studies (Podsakoff & Organ, 1986). In order to avoid CMV, the following procedures were taken. First, the questionnaire stated in the first page that all survey is anonymous and there is no definite answer for each question (Podsakoff, MacKenzie, Lee,
& Podsakoff, 2003); second, the range of each scale point was different; and finally, uni-dimensionality of the measures was verified with exploratory factor analysis and confirmatory factor analysis.
Measurement
Data collection was designed through questionnaire, and the measurement for each variable are listed below. The questionnaire is as shown in Appendix: Questionnaire of Career designed by the chosen instruments of each variables, with clear instructions for the questionnaire and the demographic that were required.
Career Adaptability
To collect data on career adaptability of the participants, the study used CAAS-Short Form, including 12 items, which was developed by Maggiori et al. (2015). The 12-items scale categorized into four dimensions, which are concern, control, curiosity, and confidence. The initial Cronbach alpha for the CAAS-SF was .90, as for the reliability of the four dimensions, .81 of concern, .82 of control, .77 of curiosity, and .83 of confidence (Maggiori et al., 2015). Each dimension has the item asking such as for concern dimension “thinking about what my future will be like,” control dimension “taking responsible for my actions,” curiosity dimension “looking for opportunities to grow as a person,” and confidence dimension
“working up to my ability” with a 7-point Likert scale, where 1equals not strong and 7 equals strongest.
Risk Aversion
The scale for measuring the risk aversion in this study is the Risk Aversion Measurement which was modified by Judge, Thoresen, Pucik and Welbourne (1999), with 8 items asking the perception of risk, and one item is a reversed question “I like to play the lottery.” Risk Aversion Measurement had a Cronbach alpha of .76 (Judge et al., 1999). Each item was designed to use a 5-point Likert scale, having 1 as strongly disagree to 5 as strongly agree.
Entrepreneurial Opportunity
Opportunity consists of various possibility in the context of the individual. To measure the entrepreneurial opportunity of the individual, the study adapted the scale of opportunity identification (Ucbasaran, Westhead, & Wright, 2009). The items asked about the probability
of individual in perceiving the surrounded opportunity, as one of the items stated, “How many opportunities for creating a new business have you pursued, that is committed time and resources to, within the last 6 months?” However, the researcher considered that pursuing a career of entrepreneur is a long-term plan, while the initial items asked the individual about perceiving opportunity within the past 6 months. Thus, the researcher adapted the scale by modifying the interval of perceiving opportunity from 6 months to 2 years. The questionnaire used a 5-point Likert scale to evaluate the entrepreneurial opportunity, 1 represents strongly disagree to 5 as to strongly agree. In study of Ucbasaran et al. (2009), the scale of opportunity identification had a reliability of .72.
Entrepreneurial Intention
Plenty of measurements were developed to measure entrepreneurial intention, whereas this study focused not only on the entrepreneurs who wished to start up a firm, but also on those who had the possibility of becoming a freelancer. Hereby, the Entrepreneurial Intention Questionnaire (EIQ) was considered to be the instrument to gather data on the intention to become an entrepreneur (Liñán & Chen, 2009). The EIQ consists of the antecedents of the entrepreneurial intention and the intention itself. According to the scale developed by Linan and Chen (2009), the questionnaire had collected data in Taiwan, and found that the antecedent personal attitude could significantly predict the intention to become an entrepreneur and another antecedent, perceived behavioral control, could predict the intention too. While the antecedent, subjective norm, was found to be not significant in Taiwanese sample. The EIQ had a Cronbach alpha of .943 (Liñán & Chen, 2009). The questionnaire is set to be a 5-point Likert scale, with 1 as strongly disagree to 5 as strongly agree.
Control Variable
Age, and gender served as the control variable in this study, according to the following study and literatures.
Age.
Age is debated on its relationship to entrepreneurial intention, as scholars have different perception on it. Some said that the youngster is likely to have lower intention to be an entrepreneur (Schwarz, Wdowiak, Almer-Jarz, & Breitenecker, 2009). On the other hand, several scholars debated on this observation with studies showing that the older people are less likely to have the desire to start a new firm or be self-employed (Kautonen, Luoto, &
Tornikoski, 2010; Levesque & Minniti, 2006). Regardless the conflict on the literature of age toward the intention of entrepreneurship, as the career development in Taiwan firms is stable and fixed, the researcher assumed that younger people tend to develop entrepreneurial intention.
To collect the information of age, an item “What is your age,” was asked with the selection of age ranges.
Gender.
There is a conflict on gender toward the topic of entrepreneurial intention. Some found that males tend to have higher intention to start up a firm and females have less intention to be an entrepreneur (Schwarz et al., 2009); while others said that this might be caused by a stereotype society that females grown under (Herrington & Kew, 2016). Under the gender balanced view in the society, there is physical and mental differences between males and females in their intention to be an entrepreneur. Thus, this study assumed that gender might affect the intention of becoming an entrepreneur or freelancer. Using one item, “What is you gender,” this question asked participants to identify themselves whether they are male (coded as 1) or female (coded as 0).
Data Analysis
This study used the statistic software IBM SPSS 22.0 for Window and AMOS to analyze and examine the collected data. First, the measures were tested through the validity and reliability analysis by using the exploratory factor analysis, confirmatory factor analysis and internal consistency reliability analysis, in order to verify the construct validity and reliability of the measurement in this study before proceeding to the analysis for examining the relationship of each variables and the Hypotheses of this study. Next, the demographic data of the study was compiled with descriptive analysis. Then, the data was analyzed with Pearson correlation analysis to present the relation of each variables. Lastly, the research performed hierarchical regression analysis to examine the Hypotheses of this study.
Validity and Reliability
Since the chosen instruments of the study was adapted from the developed measures, confirmatory factor analysis was used to verify the construct validity of the measures. As the study of Jöreskog and Sörbom (1989) stated that confirmatory factor analysis should be used to ensure the validity of the measurement construct by an assessment of the fit of the data to the measurement model of this research. The sample size to examine the validity of the adopted measurement model was 295. The result of the model fit is shown in Table 3.1. Four competing models were analyzed in AMOS, in order to compare the goodness of fit of the one-factor, two-factor, three-factor, and four-factor models. As a result, the four-factor model, which was the adopted measurement model, appeared to have the best model fit. The 𝜒2/df of the measurement model was 2.13. Gefen, Straub, and Boudreau (2000) stated that 𝜒2/df shall be below the threshold of 3. RMSEA was .06, which are suggested to be below the threshold
of .08 (Steiger, 1990). In this model, GFI was .84 and CFI was .91, as suggested that GFI over .80 and CFI over .90 are preferable (Byrne, 1994); TLI and IFI of this measurement were over the value of .90 (Bagozzi & Yi, 1988; Hu & Bentler, 1999). As a conclusion, the result revealed that the adopted four-factor measurement model was appropriate in this study.
Table 3.1.
Summary of Measurement Model Fit (N = 295)
Model 𝜒2 df 𝜒2/df GFI NFI CFI RMSEA TLI IFI
One-factor Model 3689.17 377 9.78 0.41 0.33 0.35 0.17 0.30 0.36 Two-factor Model 1502.13 376 3.99 0.70 0.73 0.78 0.10 0.76 0.78 Three-factor
Model
1049.11 374 2.80 0.79 0.81 0.80 0.07 0.85 0.87
Four-factor Model 783.45 367 2.13 0.84 0.85 0.91 0.06 0.91 0.92 Note.
The One-factor Model combined items of career adaptability, risk aversion, entrepreneurial opportunity, and entrepreneurial intention into one single factor.
The Two-factor Model combined items of risk aversion, entrepreneurial opportunity, and entrepreneurial intention into one factor against career adaptability as another factor.
The Three-factor Model combined items of entrepreneurial opportunity and entrepreneurial intention into one factor against career adaptability and risk aversion as two other factors.
The Four-factor Model was the proposed model.
Figure 3.3. CFA measurement model fit with career adaptability (N= 295)
The CR and AVE of each item is listed in Table 3.2. Based on previous studies, when the value of CR exceeded 0.70 and value AVE over 0.50 (Fornell & Larcker, 1981), the measurement model is considered acceptable on the convergent validity. Thus, the CR and AVE of this measurement for entrepreneurial intention (CR = 0.95; AVE = 0.77), entrepreneurial opportunity (CR = 0.87; AVE = 0.70), and career adaptability (CR = 0.92; AVE
= 0.50) were above the required value; yet, the AVE value of risk aversion (CR = 0.74; AVE
= 0.32) was below 0.50. This seems to be a slight convergent validity issue for the variables, but scholars have stated that the construct is still considered adequate, when the AVE is above 0.40 or the CR value has exceeded 0.70 (Fornell & Larcker, 1981).
Table 3.2.
Summary of Average Variance Extracted and Composite Reliability (N = 295)
AVE CR
Entrepreneurial Intention 0.77 0.95
Career Adaptability 0.50 0.92
Risk Aversion 0.32 0.74
Entrepreneurial Opportunity 0.70 0.87