This chapter illustrates the methodology of this research, including the research framework, hypothesis, research procedure, target population, data collection, measurement, and data analysis methods. This research adopted the quantitative research approach to examine the relationships among psychological capital (PsyCap), career capital, and career success.
Research Framework
The independent variable was Psychological capital (PsyCap). PsyCap consisted of four dimensions, hope, resilience, optimism, and self-efficacy. The dependent variable was career success and was measured by objective (hierarchical Status), and subjective (career satisfaction) career success. Career capital served as the mediator on the relationship between PsyCap and career success. Career capital included three dimensions, knowing how, knowing why, and knowing whom. The research framework was shown below, see Figure 3.1.
Career Capital
Research Hypotheses
Based on previous literature, the research purpose, and research questions, the hypotheses were described as follows:
Table 3.1.
Research Hypotheses
Hypotheses Explanation
Hypothesis 1. Psychological capital has a positive effect on career success.
Hypothesis 2. Psychological capital has a positive effect on career capital.
Hypothesis 3. Career capital has a positive effect on career success.
Hypothesis 4a. Career capital mediates the relationship between psychological capital and career satisfaction.
Hypothesis 4b. Career capital mediates the relationship between psychological capital and hierarchical status.
Research Procedure
This section demonstrated the procedure that was followed for the completion of this study and the procedures were presented in Figure 3.2. Identifying a research topic based on the literature was the first step. Reviewing the relevant literature to identify viable areas of interest related to our topic was the second step. Writing the research questions and hypotheses was the following step. The research questions and hypotheses were developed based on the literature review performed before. After the research questions and hypotheses were proposed, a research framework was developed to understand the relationship between variables. According to the
literatures, a questionnaire was designed to retrieve the information that this study needed to answer the research questions and test the hypotheses. The next step was to collect the data. Once the data were gathered, data analysis, hypotheses testing and interpreting of the meaning was done. Finally, results, discussions and suggestions were discussed to answer the research questions and the purpose of this study.
Identify of the research topic
Conduct literature review
Define research questions and hypotheses
Develop research framework
Select and choose questionaire from previous literatures
Collect the data
Analyze the data
Provide results, discussions and suggestions
Figure 3.2. Research Procedure
Sample
The target population of this study was the recruiters in career fairs on university campus in Taiwan. The recruiter is an individual who works to fill job openings and finds individuals qualified for positions in businesses or organizations. The sample data were collected using self-reported and paper-based questionnaires. 530 copies of the questionnaires were distributed directly to the recruiters in campus recruitments and career fairs in Taiwan.
There were two reasons for choosing recruiters as target sample in this study.
One was that recruiters play an important role in human resource management.
Human resource recruiting is a primary function in companies. Their tasks are to find the appropriate people and place them in the right positions. The other reason was the accessibility to the participants so that the researcher can approach them in person.
In the present study, a total of 530 questionnaires were distributed to recruiters in career fairs on university campus in Taiwan. 507 questionnaires were collected. 91 cases with the majority of missing values, incomplete data, and answering in the consistently specific Likert scales were excluded. Finally, 416 questionnaires were valid for the effective response rate of 78.5%.
Data Collection
The major channel to approach recruiters was career fairs on university campus in Taiwan. In the graduation season every year from March to June, a lot of companies actively contact different universities to ask for recruitment opportunities.
Similarly, colleges also invite various companies as vendors in the career fairs. The information of career fairs on university campus such as the date and the place were acquired from the 104 official website, newspapers and magazines. For the reason of
accessibility, the researcher directly distributed questionnaires to recruiters with a convenient sampling. The recruiters received a coupon of the convenient store as an incentive for filling the questionnaires. It was a confidential study. The questionnaire did not ask respondents to fill in their personal information.
Questionnaire Design
The mature scales of each variable were adopted in the questionnaires in order to ensure the content validity of questionnaires. There were three parts in the questionnaire. The first part of questionnaire included both career capital (19 items) and career satisfaction (5 items). Based on the participants’ work experience, the total 24 items were answered by using the 7-point Likert scale ranging from (1) strongly disagree to (7) strongly agree. The second part was PsyCap which included 24 items in four dimensions. In this part, the 5-point Likert scale from (1) strongly disagrees to (5) strongly agree was adopted to indicate the level of individuals’ PsyCap. In addition, there were three reverse items in this section. The last part of questionnaire was demographic information. The participants were asked to fill in their age, tenure in current job, total tenure, gender, three hierarchical status questions (permission to delegate work, project responsibility, and official leadership position), position, education level, annual salary, and industry.
In order to verify the questionnaire content validity, this research conducted the confirmatory factor analysis (CFA) by applying the AMOS 18.0 statistical software.
This study adopted indicators had been proposed in previous researches to examine the construct validity of career capital, psychological capital, and career satisfaction.
In addition to validate the questionnaire, the questionnaire may face the problem of common method variance (CMV) (Podsakoff, MacKenzie, Lee, & Podsakoff, 2003)
since we put all variables’ scales into a questionnaire which the independent variable and dependent variable were answered by same person at same time.
As a result, the researcher did the pre-preventions and post-preventions to avoid CMV. There are three ways in pre-preventions. First, the research purpose and variables’ names did not reveal in the questionnaire content. Second, the questionnaire used both 5-point and 7-point Likert scale to measure different variables in order to reduce the CMV problem. Third, reverse items were contained in the questionnaire to avoid random answers (Podsakoff et al., 2003).
Harman’s one-factor analysis was conducted in this study as post-prevention to detect the CMV problem (Podaskoff & Organ, 1986). The general proposition of Harman’s one-factor analysis is to see if the majority of the variance can be explained by a single un-rotated factor. The CMV problem exists when the first principle factor accounted for over 50% of total variance.
As the original scales were English version and this study was conducted in Taiwan, the researcher translated the scales into Chinese. The back translation was conducted to confirm both English and Chinese version with the consistent meanings.
The researcher found a Ph.D. student who studied in the Graduate Institute of Translation and Interpretation in National Taiwan Normal University. He translated the Chinese version scale back to English. To check whether Chinese and English version are in the same meaning or not, the researcher went through the expert review to make sure the words translated into the consistent meaning. Finally, the Chinese version was formed through the discussion with advisor. The questionnaire can be seen in appendix A.
Measurement
The measurement comprises three measures: career capital, career success, and psychological capital. The details are as follows:
Career Capital
Career Capital was measured with three dimensions scale, knowing-why, knowing-why, and knowing-how. Of three dimensions of career capital, knowing why (9-items), and knowing whom (4-items) were developed by Jokinen, Brewster and Suture (2008) and their corresponded original Cronbach’s alpha scores were 0.78, and 0.77, respectively. And the knowing-how (6-items) was developed by Eby et al. (2003) and Baker and Aldrich (1996). The original Cronbach’s alpha for knowing-how was 0.87. Respondents were asked to indicate their level of agreement with the scale items.
A total of 19 items were incorporated into three of career capital scale. The research sample presented a coefficient alpha of 0.91 (See Table 4.2), indicating good internal consistency.
Knowing-why career capital was measured through nine items scale to acquire the understanding of individual values, work interests and capabilities. Sample item is
“Recognizing your own strengths and weaknesses, needs and motives.” Each item was answered using a 7-point Likert-type scale ranging from (1) strongly disagree to (7) strongly agree.
Knowing-whom career capital, a four-item scale was used to measure the development of networking skills and social networks based on the work experience.
Sample item is “‘Knowledge of people with influential power within organizations.”
It reflects the career capital perspective of the social networking developed during the work, focusing on the relationships that have some level of influential power within
organization from the relationships with other people without such power. Each item was answered using a 7-point Likert-type scale ranging from (1) strongly disagree to (7) strongly agree.
Knowing-how was assessed with a six items combining the one item of the transferability of one's job-related skills and knowledge to other employment settings developed by Baker and Aldrich (1996) and the five items career-job-related skills scale developed by Eby et al. (2003). Sample item are “I have a diversified set of job-related skills” and “My job-related knowledge and skills are easily transferable/applied to other employment settings” Each item was answered using a 7-point Likert-type scale ranging from (1) strongly disagree to (7) strongly agree.
Career Success
In this study, career success was measured by objective career success and subjective career success.
Objective career success.
Hierarchical status represented individuals’ objective career success by using three items scale developed by Abele and Spurk (2009). The three items were permission to delegate work (0 = no, 1 = yes), project responsibility (0 = no, 1 = yes), and official leadership position (0 = no, 1 = yes). These three items were summed up to represent the participants’ hierarchical status. The hierarchical status variable could vary between zero and 3. The Cronbach’s alpha reliability of hierarchical status in this study was 0.66 (See Table 4.2).
Subjective career success.
Career satisfaction was used to measure individuals’ subjective career success.
Career satisfaction was measured with five items developed by Greenhaus, Parasuraman, and Wormley (1990). The participants respond to these five items on the
7-point Likert scales ranging from (1) strongly disagree to (7) strongly agree. The sample items included ‘I am satisfied with the success I have achieved in my career’,
‘I am satisfied with the progress I have made toward meeting my overall career goals’,
‘I am satisfied with the progress I have made towards meeting my goals for advancement’, ‘I am satisfied with the progress I have made toward meeting my goals for income’, and ‘I am satisfied with the progress I have made toward meeting my goals for the development of new skills’. The five items were summed up to form the career satisfaction measure. The original career satisfaction scales reported that internal consistency is with a total Cronbach’s alpha score of 0.88 to be reliable. In this study, the internal reliability coefficient of career satisfaction was 0.79 (See Table 4.2).
Psychological Capital (PsyCap)
Psychological capital (Psycap) was defined as a core psychological factor of positivity that go beyond human and social capital to gain a competitive advantage through investment and development of ‘who you are’ (Luthans et al., 2004).
Psychological capital measurement was adopted from Psychological capital (PsyCap) questionnaire (PCQ) self-report version by Luthans et al. (2007) with 24 items in 4 dimensions which were Hope (the original Cronbach’s alpha score of 0.8), Resilience (the original Cronbach’s alpha score of 0.72), Optimism (the original Cronbach’s alpha score of 0.72), and Efficacy (the original Cronbach’s alpha score of 0.85). Each dimension included six items.
Example item of hope is “If I should find myself in a jam at work, I could think of many ways to get out of it.”; example item of resilience is “I feel I can handle many things at a time at this job.”; example item of optimism is “If something can go
wrong for me work-wise, it will.”; example item of efficacy is “I feel confident analyzing a long-term problem to find a solution.” Each dimension are in the form of statement and responses are designed on a 5-Point Likert scale ranging from (1) strongly disagree to (5) strongly agree. The high score of scale indicates that an individual have high level of psychological capital (PsyCap). The research sample presented a coefficient alpha of 0.91 (See Table 4.2), indicating good internal consistency.
Control Variables
Based on the previous literatures, it is necessary to rule age, gender and position out to get a sense of the degree attributed clearly to career success.
Age.
Generally, salary increases more or less automatically with age. Besides, older people have a higher status in organizations since they have more experiences than younger employees. Age is likely to influence career success (Kuijpers, Schyns, &
Scheerens, 2006). Age and gender are the crucial individualistic differences to affect perceptions of career outcomes (Chen, 2012). Respondents answered their age in an open-ended question.
Gender.
Betz and Hackett (1981) found that men and women differ with respect to their career choices and to what determine their career success, with women feeling more competent in typically female occupations as opposed to typically male occupations.
Furthermore, Stroh, Brett and Reilly (1992) indicated that man and women’s career experiences are quite different, which had implications for understanding the relationships between the predictor and career success. In this research, the dummy variable was created to analyze the nominal variable. Hence, female was code as 0
and male was code as 1.
Position.
Referring to the indicators of objective career success, the present position in an organization has implications for salary and status (Forret & Dougherty, 2004).
Respondents self-rated their position in the organizations using following categories:
1. Employee, 2. Supervisor, 3. Middle level manager, and 4. Top level manager.
Job tenure.
Prior researches indicated that job tenure was positively related to career outcome (Judge et al, 1995; Stroh et al., 1992). The result showed that individual with longer job tenure may have developed knowledge, skills in their position and acquired valuable experiences. Respondents indicated their number of year of work experience.
Reliability and Validity
Reliability and Validity analysis aim to ensure the internal consistency and stability of the instruments. In the reliability analysis, Cronbach’s alpha is the most widely adapted. Cronbach’s alpha presents a summary measure of the inter correlations that occur among a set of items. Considering reliability, Cronbach’s alpha higher than 0.7 is acceptable (Guilford, 1965; Nunnally, 1978). For the validity analysis, confirmatory factor analysis (CFA) was conducted to make sure the content validity of the measurement.
Pilot Test
Before collecting the data, the researcher conducted pilot test to ensure the wording and the meaning of items in the questionnaires were appropriate that respondents can understand it. Also, reliability of the questionnaire was measured in the pilot test. 56 recruiters in career fairs on university campus in Taiwan completed
the questionnaires in the pilot test, and these 56 pilot test samples were excluded from the final survey. Table 3.2 presented the reliability of three variables and its dimensions in the pilot test. As shown in Table 3.2, Cronbach’s alpha coefficient ranged from 0.70 to 0.95 which indicated reasonable to a good consistency of the items’ set in measuring the research variables. The Cronbach’s alpha for career capital was 0.95 and its dimensions of knowing-how was 0.82, knowing-why was 0.93, knowing-whom was .86. The reliability for psychological capital was 0.95. For the Cronbach’s alpha of four dimensions of PsyCap, hope, resilience, optimism, and self-efficacy were 0.9, 0.87, 0.79, and 0.83, respectively. Lastly, the Cronbach’s alpha for career satisfaction was 0.95 and hierarchical status was 0.70.
Table 3.2.
Reliability Analysis of Pilot Test (n=56)
Variables and Dimensions Number of items Cronbach’s alpha
1. Career Capital 19 .95
1-1. Knowing-how 6 .82
1-2. Knowing-why 9 .93
1-3. Knowing-whom 4 .86
2. Psychological Capital 24 .95
2-1. Hope 6 .90
2-2. Resilience 6 .87
2-3. Optimism 6 .79
2-4. Self-efficacy 6 .83
3. Career Satisfaction 5 .95
4. Hierarchical Status 3 .70
Confirmatory Factor Analysis (CFA
)Before testing the hypotheses, confirmatory factor analysis (CFA) was performed to evaluate the distinctiveness of the measures in this study by using AMOS 18.0.
Because the chi-square (χ2) test is sensitive to sample size, the overall model fit was also examined by various fit indices, including root-mean-square error of approximation (RMSEA), Tucker–Lewis non-normed index (TLI), incremental fit index (IFI), and the comparative fit index (CFI). A good model fit was shown when RMSEA was below 0.08 (Browne & Cudeck, 1993) and TLI, IFI and CFI scores are above 0.90 (Byrne, 2001).
The results of CFA presented in Table 3.3. The value of chi square divided by degree of freedom was below 3 (Career Capital = 2.97; PsyCap= 2.37; Career Satisfaction = 1.31). The model fit was good when RMSEA was below or equal to 0.08 (Career Capital = 0.07; PsyCap= 0.06; Career Satisfaction = 0.03), the CFI (Career Capital = 0.91; PsyCap= 0.92; Career Satisfaction = 0.97), NFI (Career Capital = 0.88;
PsyCap= 0.87; Career Satisfaction = 0.99), TLI (Career Capital = 0.92; PsyCap= 0.98;
Career Satisfaction = 0.99), IFI (Career Capital = 0.89; PsyCap= 0.91; Career Satisfaction = 0.99), and the GFI score (Career Capital = 0.89; PsyCap= 0.89; Career Satisfaction = 0.99) was above 0.90. The NFI, IFI, and GFI value of career capital and the NFI and GFI scores of PsyCap were slightly below 0.90, but it still in the accepted range. However, with regards to objective career success, hierarchical status, the number of three parameters exactly equals to the number of known values which caused zero degree of freedom. It meant that the latent variable, hierarchical status with three items had a perfect just-identified model. Therefore, there was no need to conduct the CFA of hierarchical status.
In conclusion, the three variables, career capital, PsyCap, and career satisfaction,
all reported good fit indexes that ensured the validity of the research.
Table 3.3.
Results of Confirmatory Factor Analyses (n=416) Model fit indices
χ² df χ²/df RMSEA CFI NFI TLI IFI GFI 1.Career Capital 433.95 146 2.97 .07 .91 .88 .92 .89 .89
2. PsyCap 575.23 243 2.37 .06 .92 .87 .98 .91 .89
3.Career
Satisfaction 5.25 4 1.31 .03 .99 .99 .99 .99 .99
Note: PsyCap=Psychological capital RMSEA=Root Mean Square Error of Approximation;
CFI=Comparative Fit Index; NFI=Normed Fit Index; TLI=Taker-Lewis Index; IFI=Incremental Fit Index; GFI=Goodness-of-fit index.
Harman’s One-Factor Test
In the post-prevention of common method variance (CMV), Harman’s one-factor test (Podsakoff et al., 2003) was performed to detect potential bias. Since the data were self-reported and collected by the same questionnaire at the same time.
Podsakoff et al. (2003) stated that if a single factor emerged or one general factor account for the majority of all the dependent and independent variables means that the CMV problem exists in the research. In this study, all the 48 items were entered into the factor analysis and then the un-rotated factor solution was examined to determine the number of variances of the variables. The result revealed 10 factors with the eigenvalue greater than 1.0, and accounted for 62.38% of variance. The first principle factor accounted for 30.29% of total variance, which was less than 50%. Hence, the result showed that the items did not load on a single factor and minimize common method bias concern.
Data Analysis
In this study, the data was analyzed by using SPSS v18.0 and AMOS 18.0 statistical software. The analysis method included confirmatory factor analysis (CFA), descriptive statistics, Pearson Correlation, and hierarchical regression analysis.
Before testing the hypotheses, confirmatory factor analysis (CFA) was conducted to test the theoretical measurement structure and examine how well the data set fits the measurement structure used in this research by operating AMOS 18.0. This study used descriptive statistic to provide variables’ description and samples’ profile information and measure the mean, standard deviations, variance and frequency distributions for demographic characteristics. The Pearson Correlation was performed to provide a preliminary view of relationship between pairs of variables. Both independent variable and dependent variable were continuous variable, thus, hierarchical regression analysis was adopted to test hypotheses and the mediating effect of career capital in this study.
Based on Baron and Kenny’s (1986) recommendations, they defined four steps to
Based on Baron and Kenny’s (1986) recommendations, they defined four steps to