**4. Materials and methods**

#### **4.1 Respondents**

Participants were a convenience sample of 214 high school teachers (grade 11 and 12), working in the Eastern Cape Province of South Africa, accessed over a 10-month period. The sample by gender consisted of 53% females and 47% males; by race, the majority (95%) were from the black population and only 5% were Coloureds. About 65% were aged between 30 and 40 years, and most of the sample (70%) had over 10 years of work experience.

#### **4.2 Measures and outcomes of reliability tests**

Given that three main constructs existed in this study, this section reports on the measures used for these constructs. Career adaptability was measured through four subscales on a measure originally designed by Savickas and Porfeli [43]. The measure for career adaptability had four subscales, namely, concern, control, curiosity, and confidence. Each of the mentioned subscales is made up of six items, each measured on a five-point Likert scale ranging from 1 = not strong to 5 = strongest. We performed reliability tests of the career adaptability scale and found associated Cronbach alpha coefficients as follows: concern (0.735), control (0.718), curiosity (0.742), and confidence (0.739).

The presence of a calling was measured through a scale consisting of two items measured on a five-point Likert scale (1 = strongly disagree to 5 = strongly agree), with a Cronbach alpha of 0.766. The scale used to measure this was the Brief Calling Scale [44]. The scale consists of two statements: "I have a calling to a particular kind of work", and, "I have a good understanding of my calling as it applies to my career." Employee health consisted of two measures of health, namely physical and mental health. The physical health scale with a Cronbach alpha coefficient score of 0.795 was developed from a previous work [45]. The physical health scale is made up of six items measured on a five-point Likert scale with scores ranging from 0 = none of the time to 4 = all of the time. Finally, the second indicator of employee health was mental health taken from a previous study [46] and yielding a Cronbach alpha coefficient score of 0.751. The mental health scale is made up of six items measured on a five-point Likert scale ranging from 0 = none of the times to 4 = all of the times. Based on the analysis we conducted, all yielded Cronbach alpha coefficient scores were above the recommended threshold of 0.7 [47, 48]. A Cronbach coefficient

*The Influence of Career Adaptability on Career Calling and Health of Teachers DOI: http://dx.doi.org/10.5772/intechopen.93563*

equal to or above 0.7 indicates that a scale has internal consistency. This means that the items within a scale are correlated and they all speak to the same theme that is being investigated. In other words, the scales used to measure the variables utilized in this study can be relied upon, as they are reliable.

#### **4.3 Procedure**

Ethical clearance was granted through the University of Fort Hare Research Ethics Committee (UREC), with a certificate number CHI001 (Project). Participants consented to take part in the study and were approached over a 10-month period at teaching and learning summits organized within the Eastern Cape Province of South Africa.

#### **4.4 Data analysis**

Given the presented hypotheses, simple linear regression analysis was utilized to measure the influence of the presence of a calling on employee mental and physical health, that is H1 and H2. To measure the moderating effect of career adaptability on employee mental (H3) and physical health (H4), hierarchical regression analysis was undertaken. The Statistical Package for Social Sciences (SPSS) version 25 was used to perform the statistical analysis.

## **5. Results**

#### **5.1 Presence of calling and mental health**

To test the hypothesis stating that the presence of calling predicts employee mental health, the independent variable (IV) data, that is the presence of calling and the dependent variable (DV), that is employee mental health were checked to determine whether they adhered to the assumptions of normality. The Kolmogorov-Smirnov and the Shapiro-Wilk tests were performed and the results revealed that the IV and DV data violated the assumptions of normality, given *p* = .001. With this result, the simple linear regression was performed and the bootstrapped confidence intervals and their associated significance levels were requested. These do not rely on assumptions of normality and homoscedasticity.

Simple linear regression analysis output revealed a moderate positive correlation between presence of calling and employee mental health, *r* = 0.406. The presence of calling further accounted for 16.4% of the variation in employee mental health, *R*2 = 0.164. As much as there are other factors accounting for approximately 84% of the variation in employee mental health not measured by this model, the results imply that if we are to better understand the concept of employee mental health, it is worth paying attention to the presence of calling as a factor. Further, the results also revealed the *F*—ratio which seeks to explain if the model is better at predicting employee mental health as opposed to making use of the mean value. The *F*-value = 41.732 and significant *p* = .001 were observed. This result informed us that the study's model is significantly better at predicting employee mental health as opposed to relying on the mean value. **Table 1** provides a summary of the discussed results.

Parameters of the model estimates were observed, *b*0 = 20.157 and *b*1 = 0.438. This result informed us that given a unit change in the presence of calling, there will be an increase in the score of employee mental health equivalent to 0.438 units. More importantly, this positive change is significant at *p* < .001. Given this result,


**Table 1.**

*\*Significant fit.*

*Simple linear model fit and summary for the presence of calling on mental health.*

we failed to reject the hypothesis stating that the presence of calling predicts employee mental health as there is enough evidence to substantiate this claim. **Table 2** provides a summary of the mentioned results.

#### **5.2 Presence of calling and physical health**

The study's second hypothesis sought to establish whether the presence of calling predicted employee physical health. The IV and DV data were subjected to normality tests, namely the Shapiro-Wilk test and the Kolmogorov-Smirnov test, which were both observed to be below 0.05. This result indicates that the data do not adhere to the assumptions of normality. As a result, to test the hypothesis, a simple linear regression analysis was performed and bootstrapped confidence intervals, and their associated significant levels were requested as these are not affected by the assumptions of normality and homoscedasticity.

Simple linear regression analysis results revealed a very weak positive correlation between presence of calling and employee physical health, *r* = 0.122. Further, the results revealed that presence of calling accounted for just 1.5% of the variation in employee physical health, *R*<sup>2</sup> = .015. This informed us that approximately 98% of the variation in employee physical health is accounted for by other factors not measured by this model. The *F*-ratio = 3.22 and nonsignificant, *p* = 0.074 indicate that this model is not better at predicting employee physical health as opposed to the use of the mean value. **Table 3** summarizes this discussion.

Parameter estimates of the model, *b*0 = 27.160 and *b*1 = 0.119, were observed. From these parameters, we learnt that a unit change in the presence of calling results in a positive but nonsignificant change of 0.119 units in employee physical health, *p* = 0.074. Given this result, we therefore rejected the hypothesis stating that presence of calling predicts employee physical health as there is no sufficient evidence to support the hypothesis. These results are outlined in **Table 4**.

#### **5.3 Linear model predictors of mental health**

Hierarchical regression analysis was performed with career adaptability entered as the moderating variable, presence of calling as the explanatory variable, and employee mental health as the outcome variable. The results indicated a significant negative relationship between presence of calling and employee mental health, *b* = −0.052, 95% CI [−0.074, −0.030], *t* = −4.678, *p* = 0.001. The conditional effects of the focal predictor at values of the moderator output were observed which revealed that given a low value of career adaptability = −4.622, a relatively strong

*The Influence of Career Adaptability on Career Calling and Health of Teachers DOI: http://dx.doi.org/10.5772/intechopen.93563*


*\* Significant fit.*

#### **Table 2.**

*Parameter estimates for the presence of calling on mental health.*


*Note: Independent variables: constant, calling. Dependent variable: physical health.*

#### **Table 3.**

*Simple linear model fit and summary for the presence of calling on physical health.*


*Note: Independent variables: constant, calling. Dependent variable: physical health. \*Significant fit.*

#### **Table 4.**

*Parameter estimates for the presence of calling on physical health.*

and significant positive relationship between presence of calling and employee mental health does exist, *b* = 0.612, 95% CI [0.437, 0.786], *t* = 6.912, *p* = 0.001, and when the value of career adaptability increases to −1.622, a relative weak but significant positive relationship between presence of calling and employee mental health does exist, *b* = 0.455, 95% CI [0.321, 0.589], *t* = 6.699, *p* = .001. The results also revealed that with a further increase in career adaptability score, to be precise, given career adaptability = 4.379, although significant and positive, the relationship between presence of calling and employee mental health further weakens, *b* = 0.142, 95% CI [0.004, 0.280], *t* = 2.026, *p* = 0.04.

The Johnson-Neyman method revealing the significance level regions on the relationship between presence of calling and employee mental health at the values of career adaptability further confirmed the above results indicating that the relationship between presence of calling and employee health is stronger, positive, and significant at the weakest score of career adaptability, to be precise, career adaptability = −14.622, *b* = 0.134, 95% CI [0.767, 1.502], *t* = 6.082, *p* = 0.001. Conversely, at the highest score of career adaptability, that is, career adaptability = 18.379, a

very weak and significant negative relationship between presence of calling and employee mental health is observed, *b* = −.589, 95% CI [−0.986, −0.193], *t* = −2.929, *p* = 0.004. Given this result, it is evident that career adaptability does moderate the relationship between presence of calling and mental health as shown by an inverse negative relationship, *b* = −0.052. However, this finding does not support the hypothesis stating that career adaptability moderates the relationship between presence of calling and employee mental health such that when career adaptability is high, the relationship between presence of calling and employee mental health is stronger, and as a result, we rejected the stated hypothesis as there is no evidence to support it.

#### **5.4 Linear model predictors of physical health**

Hierarchical multiple regression analysis was also performed with career adaptability as a moderating variable, presence of calling as an explanatory variable and employee physical health as the outcome variable. The results revealed a significant negative relationship between presence of calling and employee physical health, *b* = −0.033, 95% CI [−0.056, −0.010], *t* = −2.873, *p* = 0.005. Examining the conditional effects of the focal predictor at values of the moderator, the results indicated that at the lowest value of career adaptability, to be precise, career adaptability = −4.622, a relatively strong significant positive relationship between presence of calling and employee physical health is observed, *b* = 0.199, 95% CI [0.019, 0.379], *t* = 2.175, *p* = 0.03. However, when the value of career adaptability increases to −1.622, a relatively weak nonsignificant positive relationship between presence of calling and employee physical health is observed, *b* = 0.099, 95% [−0.039, 0.238], *t* = 1.418, *p* = 0.16. The relationship between presence of calling and employee physical health further weakens when career adaptability increases, that is, career adaptability = 4.379, resulting in a nonsignificant negative relationship between presence of calling and employee physical health, *b* = −0.099, 95% CI [0.242, 0.043], *t* = −1.372, *p* = 0.172.

The Johnson-Neymar method was further observed to determine the significance regions on the relationship between presence of calling and employee physical health at the values of the moderator. At the lowest value, that is, career adaptability = −14.622, a relatively strong and significant positive relationship between presence of calling and employee physical health is observed, *b* = 0.529, 95% CI [0.150, 0.908], *t* = 2.753, *p* = 0.006. As the value of career adaptability increases, to 6.661, a relatively weak and significant negative relationship between presence of calling and employee physical health is observed, *b* = −0.175, 95% CI [−0.350, 0.000], *t* = −1.971, *p* = 0.05 but when the value of career adaptability is increased further, the relationship between presence of calling and employee physical health further weakens, for example, at the highest value of career adaptability, that is, 18.379, a significant negative relationship between presence of calling and employee physical health is observed, *b* = −0.562, 95% CI [−0.971, −0.153], *t* = −2.71, *p* = 0.007. From this finding, we also observed that career adaptability moderates the relationship between presence of calling and physical with the interaction revealing an inverse relationship, *b* = −.033. This finding does not support the study's hypothesis stating that career adaptability moderates the relationship between presence of calling and employee physical health such that when career adaptability is high, the relationship between presence of calling and employee physical health is stronger. Therefore, we rejected the stated hypothesis as there is no evidence to substantiate this claim.
