Descriptive And Inferential Statistics Essay Results.

Participant Demographics

In this section, descriptive and inferential statistics results are reported. The descriptive statistics plays an important role in providing an insight regarding the different variables in the study. For instance, it provides us with the frequency distribution of participant demographics as well as the responses necessary for answering the research objectives. Hence, the analysis acts as a pillar through which inferential statistics is performed. In particular, the results including tables, graphs (pie chart, bar graphs, histogram and column graphs) are reported. Inferential statistics on the other acts as key pillar in answering the research questions and objectives. This involved different analytic technique specific to the objective under study. The results of acquired are discussed below.

Descriptive statistics

Descriptive statistics were analyzed in two phases, one for the demographics which was key in understanding participant characteristics and for variables of interest that gives insight for the study variables. These are shown below.

Age

The first demographic data analyzed was on age where the frequency bar graph was used. This is displayed in figure 1 below.

 

Fig 1. Bar graph for age count

From the above table, most respondents were of the age 41 – 50 years (n = 118) same as those between 31 and 40 years. These were followed by the 21-30 years’ group (n = 92), 51 – 60 years (n = 65) and 60 years and above (n = 7). This shows that majority of the respondents were on their active working period and were hence relevant for participation in the study where workplace experience was required.  

Position

The next step involved analyzing frequency of position. This was done using pie chart and this is represented in the figure 2 below.

 

Fig 2. Pie chart for position by participant

Majority of the respondents (n = 292, 73.00%) were not managers followed by a category that had other managerial roles (n = 74, 19.00%), trainee (n = 20, 5.00%), mid-level manager (n = 10, 2.00%) and finally top level manager (n =4, 1.00%). This is an indication that most respondents were not on top hence could be subjected to different wellbeing, anxiety and perception towards work environment, a condition perfect for this study topic. For instance, managers are at times reserved from speaking negative about the working condition of the banks, a factor that prevents them from speaking what they truly feel thus affecting the study results, a factor that’s different for the non-managers.

Relationship status

The status of relationship is impactful in terms of wellbeing of an individual and thus, reporting this status for the participant was done using pie chart as displayed in figure 3 below. 

 

Fig 3. Pie chart for relationship status

From the above graph, most participants (n = 286, 71.00%) were married compared to the 114 (29.00%) that were not married. This acted as a distinguisher in establishing the different levels of feeling well and anxiety levels.

Education level

Education level is a key pillar in the understanding of work conditions and workaholic state. For instance, the more learned usually have the required skill and expertise to achieve more. It is also in order to get the more learned in filling the questionnaire with minimum supervision. This variable was presented using column graph as shown in figure 4 below.

Insights on Frequencies of Age, Position, Relationship Status, and Education Level

 

Fig 4. The level of education

Based on the above figure, most of the respondents were of masters’ degree qualification (n = 227) followed by Bachelor’s degree (n = 164) and finally PhD (n = 9). This cohort was composed by an all learned working group that had exhaustive knowledge in regards to the questions of study and thus gave the desired results. This group must have also underwent different workaholic influencing factors important for this study.

Gender

The last variable in this category is gender. Establishing the gender distribution is important in understanding the fairness in sampling selection. Gender frequency is presented using the table below.

Table 1. Gender frequency

Row Labels	Frequency	Percentage
Female	200	50.00%
Male	200	50.00%
Grand Total	400

Based on the table above, there were equal number of males and females (n = 200, 50.00% each). This shows that there was no sampling bias in terms of gender, a factor that is necessary for achieving the desired study results.

Workaholic status

Being a variable of interest, there was a need to display the distribution frequency; this was done using the pie chart below.

 

Fig 5. The frequency chart of workaholic state

Majority of the respondents (n = 357, 89.00%) were non – workaholic compared to 43 (11.00%) that were workaholic. It can therefore be concluded that most respondents were not addicted to work. This result is important in explaining the variation of feelings including wellbeing of respondents with respect to their places of work.

Anxiety levels

 This was also represented using pie chart considering that the data was categorical in nature measured on a nominal scale. The resulting graph is displayed below.

 

Fig 6. Anxiety level pie chart

From the above graph, most of the respondents (n = 350, 87.00%) were of low anxiety compare to 50 (13.00%) that were of moderate anxiety. This is a representation of the varying psychological feeling that is key in explaining the varying workaholic state of the respondents.

Wellbeing and perceived work environment

Summary statistics tests were carried out to display the insight of wellbeing and environmental scores. Other than insights, the distribution of the two variables was established, a factor that was key to the choice of parametric or nonparametric test. The values are presented in the table 2 below.  

Table 2. Summary statics test results

Statistics
		Perceived Work Environment Score	Wellbeing
N	Valid	400	400
	Missing	0	0
Mean		143.97	9.38
Median		146.00	5.00
Std. Deviation		29.32	8.069
Skewness		-.148	0.630
Std. Error of Skewness		.122	0.122
Minimum		76	0
Maximum		210	29

Based on the above table, the average score of the perceived work environment was 143.97 (SD = 29.32) whereas that of the wellbeing was 9.38 (SD = 8.07). While the lowest and highest scoring participants in terms of perceived work environmental were correspondingly 76 and 210 in terms of values, the wellbeing scores were 0 for lowest and 29 for highest. The distribution of perceived work environment was almost normal with a skewness value of -.15; this is displayed in the graph below.

 

Fig 7. Histogram for perceived work environment scores

The graph further confirms the normality of the perceived work environment scores.

Differently, wellbeing was more skewed positively (.63) and this is displayed using the histogram below.

Workaholic State and Anxiety Level

 

Fig 8. Histogram for wellbeing scores

As confirmed, some scores were skewed to the right showing that most people recorded lower wellbeing scores.

Inferential statistics

Different sets of inferential statistics were performed to answer the research objectives. These are discussed below.

To establish if there is significant difference between workaholic & non-workaholic group over anxiety, psychological well-being, and the perceived work environment

This objective was tested using three different analyses corresponding to anxiety, psychological wellbeing and work environment scores. These are discussed below.

Relationship between workaholic and anxiety

The anxiety condition (moderate and low) being categorical (nominal) same as workaholic state (workaholic vs non-workaholic) involved the use of Pearson Chi-square test along with contingency table. This was done at 5% level of significance and the results displayed below.

Table 3. Contingency table

Anxiety Level * Workaholic Cross tabulation
Count
		Workaholic		Total
		Workaholic	Non-workaholic
Anxiety Level	Low Anxiety	0	350	350
	Moderate Anxiety	42	8	50
Total		42	358	400

From the above table, there is displayed that majority of the workaholics had moderate anxiety (n = 42) compared the non-workaholics who had the majority with low anxiety (n = 350). The significance of the distribution is displayed in the corresponding table below.

Table 4. Pearson Chi-square results

Chi-Square Tests
	Value	df	Asymptotic Significance (2-sided)	Exact Sig. (2-sided)	Exact Sig. (1-sided)
Pearson Chi-Square	328.492a	1	.000
Continuity Correctionb	319.614	1	.000
Likelihood Ratio	224.779	1	.000
Fisher's Exact Test				.000	.000
N of Valid Cases	400
a. 0 cells (0.0%) have expected count less than 5. The minimum expected count is 5.25.
b. Computed only for a 2x2 table

There was displayed a significance in the association between the anxiety levels and the workaholic state of an individual (χ2 (1) = 328.49, p = .00) thus leading to the rejection of the corresponding null hypothesis. This leads to the conclusion that the workaholics had a significantly different level of anxiety (moderate) compared to the non-workaholics.

Relationship between workaholic and psychological wellbeing

Testing this section of the objective involved two variables, one continuous (wellbeing) dependent variable and one categorical (workaholic state) independent variable. Therefore, displaying two samples, the corresponding tests are either parametric t-test or nonparametric Mann Whitney U test. Hence, to arrive at the best test, a Kolmogorov Smirnov test of normality was carried out to establish data normality. This was done under the null hypothesis that the data categories were normally distributed and this was tested at 5% level of significance. The resulting table is presented below.

Table 5. Normality test result 

Tests of Normality
	Workaholic	Kolmogorov-Smirnova			Shapiro-Wilk
		Statistic	df	Sig.	Statistic	df	Sig.
Score	Workaholic	.229	42	.000	.836	42	.000
	Nonworkerholic	.267	358	.000	.853	358	.000
a. Lilliefors Significance Correction

Based on the above table, the Kolmogorov – Smirnov p-value is <.001 for both workaholic and non-workaholic states of wellbeing. Hence, the null hypothesis is rejected for both and a conclusion is made that the two categories of wellbeing scores were not distributed normally. Hence, the best test for the hypothesis is Mann Whitney U test. This was done and the following ranks and Z-test results reported. 

Table 6. Ranks

Ranks
	Workaholic	N	Mean Rank	Sum of Ranks
Score	Workaholic	42	375.35	15764.50
	Non workaholic	358	179.99	64435.50
	Total	400

Table 7. Test results

Test Statisticsa
	Score
Mann-Whitney U	174.500
Wilcoxon W	64435.500
Z	-10.455
Asymp. Sig. (2-tailed)	.000
a. Grouping Variable: Workaholic

Based on the above results, the workaholic group ranked high (M = 375.35) on the wellbeing scores compared to the non-workaholic category (M = 179.99). This difference was significant at 5% level of significance (Mann-Whitney U = 174.50, p = <.001) thus leading to the rejection of the null hypothesis that there was no evident relationship between the workaholic and non-workaholic wellbeing scores. Hence, it is concluded that the level of wellbeing was dependent of the workaholic state of the respondents.

Perceived Work Environment Scores

Testing this section of the objective involved two variables, one continuous (perceived work environment) dependent variable and one categorical (workaholic state) independent variable. Therefore, displaying two samples, the corresponding tests are either parametric t-test or nonparametric Mann Whitney U test. Hence, to arrive at the best test, a Shapiro Wilk test of normality was carried out to establish data normality. This was done under the null hypothesis that the data categories were normally distributed and this was tested at 1% level of significance. The resulting table is presented below.

Table 8. Normality test result 

Tests of Normality
	workaholic	Kolmogorov-Smirnova			Shapiro-Wilk
		Statistic	df	Sig.	Statistic	df	Sig.
PWE	Workaholic	.087	43	.200*	.982	43	.745
	Nonworkaholic	.069	357	.000	.984	357	.000
*. This is a lower bound of the true significance.
a. Lilliefors Significance Correction

Based on the above table, the Shapiro Wilk p-value is .75 for workaholic and .00 for the non-workaholic states of perceived work environment. Hence, the null hypothesis is accepted for both and a conclusion is made that the two categories of perceived work environment scores were distributed normally. Hence, the best test for the hypothesis is Independent sample t-test. To ascertain which between equal and unequal variance, a Levene’s test of variance homogeneity was carried out under the null hypothesis that the variances are equal. The resulting p-value was .81 which is greater than 5% level of significance; the null hypothesis is accepted as true and the variances are assumed to be equal. An independent sample t-test was then conducted and the results displayed below.

Table 9. Summary statistics results

Group Statistics
	workaholic	N	Mean	Std. Deviation	Std. Error Mean
PWE	Workaholic	43	143.5116	28.98306	4.41988
	Nonworkaholic	357	144.0252	29.39946	1.55598

Table 10. Independent sample t-test results

Independent Samples Test
		Levene's Test for Equality of Variances		t-test for Equality of Means
		F	Sig.	t	df	Sig. (2-tailed)	Mean Difference	Std. Error Difference	95% Confidence Interval of the Difference
									Lower	Upper
PWE	Equal variances assumed	.057	.812	-.108	398	.914	-.51358	4.73866	-9.82951	8.80235
	Equal variances not assumed			-.110	52.960	.913	-.51358	4.68576	-9.91220	8.88504

From the test results above, the average perceived work environment was highest for the non-workaholic group (M = 144.03, SD = 29.40) compared to the workaholic group (M = 143.51, SD = 28.98). This difference, however, was not significant at 5% level of significance (t (398) = -.11, p = .91) thus leading to the failure to reject the null hypothesis. This led to the conclusion that there is no evident relationship between perceived work environment scores and the workaholic state of an individual.

Based on the above test results, it can be concluded that the workaholic state of the respondents was significantly associated with the anxiety levels where moderate anxiety is common with individuals who are non-workaholic; and the wellbeing scores where the rank is high for the workaholics. However, the workaholic state of an individual has no evident relationship with the perceived work environment scores suggesting that the workaholics and the non-workaholics had similar scores of the same.

The two variables under study gender (male and female) and workaholic state are categorical variables of nominal measurement. Therefore, it is a clear sign that a nonparametric Pearson Chi-square test is the most appropriate. This was done at 5%v level of significance alongside clustered column graph. The first result corresponds to the graph and this is displayed below.

 

Fig 9. Clustered column graph for gender by workaholic state

Based on the graph above, it is observed that non-workaholic proportions were the highest for both males and females. The proportion was almost equal for the two workaholic categories across gender. The significance in the frequency difference is displayed using the table below.

Table 11. Pearson Chi-square test results

Chi-Square Tests
	Value	df	Asymptotic Significance (2-sided)	Exact Sig. (2-sided)	Exact Sig. (1-sided)
Pearson Chi-Square	.026a	1	.872
Continuity Correctionb	.000	1	1.000
Likelihood Ratio	.026	1	.872
Fisher's Exact Test				1.000	.500
N of Valid Cases	400
a. 0 cells (0.0%) have expected count less than 5. The minimum expected count is 21.50.
b. Computed only for a 2x2 table

From the test results, there is no significant relationship between gender and the workaholic state at 5% level of significance (χ2 (1) = .03, p = .87) thus leading to the failure to reject the null hypothesis that the workaholic state of the respondents is independent of their gender. Hence, it is concluded that one being a workaholic or not was not impacted by sex.

Gender differences on workaholic & non-workaholic groups over anxiety, psychological well-being, and the perceived work environment

Testing the impact of gender and workaholic conditions on the worker anxiety levels, wellbeing and work perceived work environment involved the use of Pearson Chi-square and univariate analysis. In the Univariate ANOVA test, gender was the random factor and workaholic state as the fixed factors and the corresponding dependent variables were wellbeing and perceived work environment. This allowed for testing the individual influences of gender and workaholic condition impact on the dependent variables as well as their interaction effect. Specifically, the established relationships and interaction effect answered the above study objectives.

The relationship between gender and anxiety levels based on the workaholic state of individuals

The two variables under study gender (male and female) and anxiety levels across the workaholic state are categorical variables of nominal measurement. Therefore, it is a clear sign that a nonparametric Pearson Chi-square test is the most appropriate. This was done at 5%v level of significance alongside clustered column graph. The first result corresponds to the graph and this is displayed below.

 

Fig 10. Clustered column graph for gender by anxiety levels

Based on the graph above, it is observed that low level anxiety proportions were the highest for both males and females. The proportion was almost equal for the two gender categories across the workaholic state. The significance in the frequency difference is displayed using the table below. 

Table 12. Pearson Chi-square test results

Chi-Square Tests
Workaholic		Value	df	Asymptotic Significance (2-sided)	Exact Sig. (2-sided)	Exact Sig. (1-sided)
Workaholic	Pearson Chi-Square	.c
	N of Valid Cases	43
Nonworkaholic	Pearson Chi-Square	.151d	1	.697
	Continuity Correctionb	.000	1	.994
	Likelihood Ratio	.152	1	.697
	Fisher's Exact Test				.723	.497
	Linear-by-Linear Association	.151	1	.698
	N of Valid Cases	357
Total	Pearson Chi-Square	.091a	1	.762
	Continuity Correctionb	.023	1	.880
	Likelihood Ratio	.091	1	.762
	Fisher's Exact Test				.880	.440
	Linear-by-Linear Association	.091	1	.763
	N of Valid Cases	400
a. 0 cells (0.0%) have expected count less than 5. The minimum expected count is 25.00.
b. Computed only for a 2x2 table
c. No statistics are computed because Anxiety Level is a constant.
d. 2 cells (50.0%) have expected count less than 5. The minimum expected count is 3.49.

From the test results, there is no significant relationship between gender and the anxiety levels at 5% level of significance (χ2 (1) = .09, p = .76) thus leading to the failure to reject the null hypothesis that the anxiety level of the respondents is independent of their gender irrespective of the workaholic condition. Hence, it is concluded that one being a low or middle level anxious was not impacted by sex across the workaholic and non-workaholic categories.

Having carried out the Univariate analysis test, the resulting values are presented in the table below.

Table 13. Univariate analysis results

Tests of Between-Subjects Effects
Dependent Variable:   Wellbeing
Source		Type III Sum of Squares	Df	Mean Square	F	Sig.
Intercept	Hypothesis	37345.286	1	37345.286	130964.452	.002
	Error	.285	1	.285a
Workaholic	Hypothesis	9641.643	1	9641.643	566.156	.027
	Error	17.030	1	17.030b
Gender	Hypothesis	.285	1	.285	.017	.918
	Error	17.030	1	17.030b
workaholic * Gender	Hypothesis	17.030	1	17.030	.414	.520
	Error	16272.849	396	41.093c
a.  MS(Gender)
b.  MS(workaholic * Gender)
c.  MS(Error)

Based on the above results, there is established a significant relationship between the workaholic conditions and the wellbeing scores (F (1, 396) = 566.16, p = .03). This leads to the rejection of the null hypothesis that there is no evident association between workaholic state and wellbeing of the participants. In contrast, gender had no evident association with the wellbeing scores of participants (F (1, 396) = .12, p = .92) thus leading the failure to reject the null hypothesis that gender and wellbeing participants were not evidently associated. This was the same result on the interactive impact of gender and workaholic conditions on the wellbeing scores of the participants where there was no significance (F (1, 396) = .41, p = 52). This led to the failure to reject the general hypothesis corresponding to the objectives that there are no significant gender differences on workaholic & non-workaholic groups over psychological well-being.

The gender difference on the perceived work environment between the workaholic and non-workaholic groups

Similar to the above objective, this was tested using the Univariate ANOVA test and the results displayed below.

Table 14. Univariate ANOVA test results 

Tests of Between-Subjects Effects
Dependent Variable:   PWE
Source		Type III Sum of Squares	Df	Mean Square	F	Sig.
Intercept	Hypothesis	3173792.544	1	3173792.544	501140.636	.001
	Error	6.333	1	6.333a
Gender	Hypothesis	797.674	1	797.674	.950	.508
	Error	839.414	1	839.414b
Workaholic	Hypothesis	6.333	1	6.333	.008	.945
	Error	839.414	1	839.414b
Gender * workaholic	Hypothesis	839.414	1	839.414	.972	.325
	Error	342063.344	396	863.796c
a.  MS(workaholic)
b.  MS(Gender * workaholic)
c.  MS(Error)

Based on the above table, there was not displayed a significant impact of workaholic state of the respondents on their perceived work environment scores (F (1, 396) = .01, p = .95) thus leading to the failure to reject the null hypothesis that there is no evident association between the two variables. Similarly, there no evident influence of gender on the perceived work environment by the respondents on their level of wellbeing (F (1, 396) = .95, p = .51) thus leading to the failure to reject the hypothetical statement that there is no significant impact of gender on the perceived work environment score of the participants. Finally, there is established no evident interaction impact of gender and workaholic condition of individuals on their scores in the perceived work environment (F (1, 396) = .97, p = .33). This led to the failure to reject the hypothetical statement that there are no significant gender differences on workaholic & non-workaholic groups over the perceived work environment. 

Objective Conclusion

Based on the above results, it can be concluded that the wellbeing of the respondents and their corresponding perceived work environment was affected by other factors rather than workaholic state and the interaction between gender and workaholic state. This was a similar case to the anxiety level of workers. In contrast, the wellbeing of workers is affected by their workaholic state, for instance, the workaholic and non-workaholic workers have significantly different levels of wellbeing.

Impact of work perceived work environment on Anxiety and Psychological Well-being of workaholic and non-workaholic group

Testing the above objective involved the splitting of the data into workaholic and non-workaholic categories to allow for the testing of the psychological wellbeing scores and work environment relationship. This, for the two categories of workaholic states, was tested using Pearson correlation test. On the other hand, the impact of work environment across the workaholic categories on the anxiety levels of respondents was tested using Binary Logistic regression on the original data. This is considering that the anxiety variable was categorical of binary nature. The resulting values are discussed below.

The impact of the perceived work environment score on the psychological wellbeing among the workaholic group

Testing the above sub-objective involved the use of Pearson correlation test at 5% level of significance; this resulted in the values of table 15 below.  

Table 15. Pearson Correlation test results

Correlations
		Wellbeing	Perceived work environment
Wellbeing	Pearson Correlation	1	-.092
	Sig. (2-tailed)		.559
	N	43	43
Perceived work environment	Pearson Correlation	-.092	1
	Sig. (2-tailed)	.559
	N	43	43

 Based on the above test results, there is displayed a small insignificant negative correlation between the psychological wellbeing of the participants and their respective perceived work environment scores (r = -.09, p = .56). This led to the failure to reject the corresponding null hypothesis that there is no evident correlation between wellbeing and perceived work environment among the workaholic group. Based on the small correlation, however, it can be stated that the value of wellbeing increases as the level of perceived work environment knowledge decreases.

The impact of perceived work environment score on the psychological wellbeing among the non - workaholic group

Testing the above sub-objective involved the use of Pearson correlation test at 5% level of significance; this resulted in the values of table 16 below.

Table 16. Pearson Correlation test results

Correlations
		PWE	Wellbeing
PWE	Pearson Correlation	1	-.027
	Sig. (2-tailed)		.594
	N	400	400
Wellbeing	Pearson Correlation	-.027	1
	Sig. (2-tailed)	.594
	N	400	400

Based on the above test results, just like for the workaholic category, there is displayed a small insignificant negative correlation between the psychological wellbeing of the participants and their respective perceived work environment scores in regards to the environment (r = -.03, p = .59). This led to the failure to reject the corresponding null hypothesis that there is no evident correlation between wellbeing and the perceived work environment among the non-workaholic group. Based on the small correlation, however, it can be stated that the value of wellbeing increases as the level of perceived work environment knowledge decreases.

The impact of perceived work environment on the anxiety levels of participants across the workaholic and non-workaholic groups

Testing the above hypothesis involved the use of Binary logistics regression at 5% level of significance. This resulted in the values below.

Table 17. Binary logistics regression results

Model Summary
Step	-2 Log likelihood	Cox & Snell R Square	Nagelkerke R Square
1	68.90a	0.441	0.833
a. Estimation terminated at iteration number 20 because maximum iterations has been reached. Final solution cannot be found.
Variables in the Equation
		B	S.E.	Wald	df	Sig.	Exp(B)
Step 0	Constant	-1.946	0.151	165.662	1	0.000	0.143
Variables in the Equation
		B	S.E.	Wald	df	Sig.	Exp(B)
Step 1a	PWE	0.001	0.013	0.008	1	0.929	1.001
	workaholic	-25.116	6128.968	0.000	1	0.929	0.000
	Constant	46.154	12257.936	0.000	1	0.997	1.107E20
a. Variable(s) entered on step 1: PWE, workaholic.

Based on the above results, there was a joint significant impact of perceived work environment scores and workaholic conditions of participants on their overall anxiety conditions (Wald = 165.662, p = <.001) thus leading to the rejection of the respective sub hypothesis that there is significant impact of work environment on Anxiety levels between the workaholic and non-workaholic group. The variables explained an average variation of the anxiety levels, 44.2%, as depicted by the value of the Cox & Snell R-squared (0.441). It is however evident that there was no impact of perceived work environment score (Wald (1) = .01, = .93) and workaholic state of respondents (Wald (1) = .008, p = 997) individually. Hence, it is clear that there is an interaction effect with no individual impact on the anxiety levels of the participants.

Based on the above sub tests, it is clear that perceived work environment scores and workaholic state of the respondents played no significant role in the anxiety and psychological wellbeing of individuals.

4.3.5. The relationship between job position and the anxiety, psychological wellbeing, and the perceived work environment of the participants between the workaholic and non-workaholic categories

Testing the impact of job position on anxiety and then psychological wellbeing at work of the participants was tested using Pearson Chi-Square and Univariate ANOVA tests respectively. These are discussed in the sub categories below.

The association between job position and the anxiety condition among the workaholic and non-workaholic group

Testing this was done at 5% level of significance using Pearson Chi-square test and the results displayed in the table below.

Table 18. Contingency table results

Position * Anxiety Level * workaholic Crosstabulation
Count
workaholic			Anxiety Level		Total
			Low level	Mid-level
Workaholic	Position	Trainee		8	8
		No Managerial		27	27
		Other managerial		8	8
	Total			43	43
Nonworkaholic	Position	Trainee	12	0	12
		No Managerial	260	5	265
		Mid-level managerial	10	0	10
		Top level managerial	4	0	4
		Other managerial	64	2	66
	Total		350	7	357
Total	Position	Trainee	12	8	20
		No Managerial	260	32	292
		Mid-level managerial	10	0	10
		Top level managerial	4	0	4
		Other managerial	64	10	74
	Total		350	50	400

It is observable from the above table that the among the workaholic group, there was no low levels of anxiety. This shows that in general, workaholic group composed of trainee, non-managers and other managerial duty holders were mainly of the mid-level anxiety category. On the other hand, it is shown that majority of the non-workaholic group of respondents had lower level of anxiety ranging from the trainees, non-managers, mid-level managers, top level managers and other managers. Only 5 out of 265 from the non-managers and 2 out of 64 from the other managers had mid-level anxiety. Linking to the workaholic group, then these two categories appeared to have counts of mid-level suggesting a possible relationship between job position and the anxiety levels. The significance of this test is reported using the table 19 of the Pearson Correlation results below.

Table 19. Pearson Chi-square test results 

Chi-Square Tests
workaholic		Value	df	Asymptotic Significance (2-sided)
Workaholic	Pearson Chi-Square	.b
	N of Valid Cases	43
Nonworkaholic	Pearson Chi-Square	.920c	4	.922
	Likelihood Ratio	1.375	4	.849
	Linear-by-Linear Association	.403	1	.526
	N of Valid Cases	357
Total	Pearson Chi-Square	16.532a	4	.002
	Likelihood Ratio	14.020	4	.007
	Linear-by-Linear Association	.467	1	.494
	N of Valid Cases	400
a. 4 cells (40.0%) have expected count less than 5. The minimum expected count is .50.
b. No statistics are computed because Anxiety Level is a constant.
c. 5 cells (50.0%) have expected count less than 5. The minimum expected count is .08.

Observed from the above table, there is no displayed association between workaholic category of the job group positions and anxiety levels, a factor justified by the non-existence of the low anxiety levels in this category. In contrast, there is an insignificant association between job group and the anxiety levels in the non-workaholic group (χ2 (4) = 1.38, p = .92) thus leading to the failure to reject a null hypothesis that among the non-workaholic group, there is no evident association between the job position of an individual and their level of anxiety. Considering the total, then there is established a significant association between the job position and the anxiety level of the study participants (χ2 (4) = 16.53, p = .002). This leads to the rejection of the general null hypothesis that the level of anxiety of the participants is independent of their job position. Linking this results to the contingency table 14 above, it is clear that the junior level employees, trainees, non-managers and those with other managerial jobs, significantly had a mid-level anxiety whereas those of higher ranks including mid-level managers and top level managers had significantly lower anxiety levels. Hence, the level of anxiety of the respondents varied based on their job position.

The association between job position and the level of wellbeing among the workaholic and non-workaholic group

Testing this relationship involved the use of Univariate analysis of variance with the above hypothesis tested based on the interaction between the workaholic and job position variables. The choice if the test was arrived at considering that there were two categorical variables as independent with a continuously measured dependent variable. This test resulted in the table 20 below.

Table 20. The Univariate ANOVA results for wellbeing, workaholic state and positions

Tests of Between-Subjects Effects
Dependent Variable:   Wellbeing
Source	Type III Sum of Squares	df	Mean Square	F	Sig.
Corrected Model	9862.792a	7	1408.970	34.268	.000
Intercept	14465.142	1	14465.142	351.813	.000
Workaholic	5528.549	1	5528.549	134.462	.000
Position	163.402	4	40.851	.994	.411
workaholic * Position	8.123	2	4.062	.099	.906
Error	16117.448	392	41.116
Total	61174.000	400
Corrected Total	25980.240	399
a. R Squared = .380 (Adjusted R Squared = .369)

Based on the above test result, three tests are reported. The first is on the workaholic association with the wellbeing of the respondents; this was established as significant (F (1, 392) = 134.46, p = .00) thus leading to the rejection of the null hypothesis that the wellbeing of respondents is substantially impacted by the corresponding workaholic state. The second test was on the job position impact on the wellbeing of the respondents; in this case, there was established an insignificant association between the variables (F (4, 392) = .99, p = .41) thus leading to the acceptance of the respective sub null hypothesis that the job position has no evident impact on the level of wellbeing of the respondents. The final test was on the interaction between workaholic state and the job position of the respondents. It was discovered from this test that there is no evident impact of the interaction between job position and workaholic state of participants on their levels of wellbeing (F (2, 392) = 4.06, p = .91). This led to the failure to reject the null hypothesis corresponding to this sub objective that there is no significant impact of job group level of study participants on their wellbeing among the workaholic and non-workaholic categories.

The association between job position and the perceived work environment among the workaholic and non-workaholic group

This was done using Univariate ANOVA under the hypothetical statement that there is an evident association between job position and the perceived work environment across the workaholic condition. This was done at 5% level of significance and the results presented below.

Table 21. Univariate ANOVA results

Tests of Between-Subjects Effects
Dependent Variable:   PWE
Source		Type III Sum of Squares	df	Mean Square	F	Sig.
Intercept	Hypothesis	1364263.624	1	1364263.624	2898.267	.000
	Error	6865.232	14.585	470.717a
Position	Hypothesis	2447.423	4	611.856	1.969	.135
	Error	6781.037	21.820	310.769b
workaholic	Hypothesis	175.153	1	175.153	.767	.404
	Error	2085.909	9.133	228.386c
Position * workaholic	Hypothesis	247.859	2	123.929	.143	.867
	Error	340050.843	392	867.477d
a. .694 MS(workaholic) - .112 MS(Position * workaholic) + .419 MS(Error)
b. .749 MS(Position * workaholic) + .251 MS(Error)
c. .860 MS(Position * workaholic) + .140 MS(Error)
d.  MS(Error)

From the above table, there is no significant impact of job position on the perceived work environment scores (F (1, 392) = 1.97, p = .14). Similarly, there is no interaction effect between job position and the workaholic state of the participants on their perceived wellbeing (F (2, 392) = .14, p = .87) thus leading to the failure to reject the null hypothesis that there is no evident impact of job position being that one is a workaholic or not on the perceived work environment.  

Based on the above test results, it is concluded that though the there is a significant impact of job position on the anxiety level of the respondents across the workaholic and non-workaholic states, this is not true for the impact of job position on the wellbeing of respondents across the workaholic groups and the impact of job position on the perceived work environment between the workaholic and non-workaholic categories.

The impact of age and relationship on the anxiety levels, worker wellbeing, and perceived work environment across the workaholic and non-workaholic categories

Like the fifth objective, testing this objective involved the use of Pearson Chi-square for categorical variables and Univariate ANOVA for continuous dependent variable. The tests were done at 5% level of significance and reported under the following sub headings.

The impact of age on the anxiety levels between the workaholic and non-workaholic categories

This was tested using Pearson Chi-square having two categorical variables. The Pearson Chi-square and the corresponding contingency table test results are presented below.

Table 22. Contingency table results

Crosstab
Count
		Anxiety Level		Total
		Low level	Mid-level
Age	21-30	71	21	92
	31-40	107	11	118
	41-50	104	14	118
	51-60	61	4	65
	Above 60	7	0	7
Total		350	50	400

Based on the above table, most of the respondents with level anxiety belonged to the 21-30-year-old age group followed by 41 – 50 years’ age group and the 31-40 group, a group that had the highest level of low anxiety participants. Of the five age categories, there was not recorded mid-level anxiety for individuals above 60 years old suggesting a possible association between anxiety and age. To ascertain whether the relationship exists, a Pearson chi-square test results are reported in table 23 below.

Table 23. Pearson Chi-square test results

Chi-Square Tests
	Value	df	Asymptotic Significance (2-sided)
Pearson Chi-Square	13.496a	4	.009
Likelihood Ratio	13.427	4	.009
Linear-by-Linear Association	8.951	1	.003
N of Valid Cases	400
a. 1 cells (10.0%) have expected count less than 5. The minimum expected count is .88.

From the test results, there is displayed a significant association between the age group of respondents and their respective levels of anxiety (χ2 (4) = 13.50, p = .01). This led to the rejection of the corresponding null hypothetical statement that the anxiety level of the respondents was independent of their respective age groups; It also confirms further the suggested relationship from the contingency table. This leads to the conclusion that the anxiety levels of individuals under study was substantially impacted by age group, in other words, people of different age groups recorded dissimilar levels of anxiety evidently.

The impact of relationship status on the anxiety levels between the workaholic and non-workaholic categories

This was tested using Pearson Chi-square having two categorical variables. The Pearson Chi-square and the corresponding contingency table test results are presented below. 

Table 24. Contingency table results

Crosstab
Count
		Anxiety Level		Total
		Low level	Mid-level
Relationship Status	Married	257	29	286
	Unmarried	93	21	114
Total		350	50	400

Based on the above table, while most of the respondents with low anxiety (n = 257) were the married, it is in the same category that recorded the highest frequency of mid-level. Nevertheless, in terms of proportion, the unmarried (18.42%) had the highest percentage for the mid-level anxiety compared to the married (10.14%) old suggesting a possible association between anxiety and age. To ascertain whether the relationship exists, a Pearson chi-square test results are reported in table 25 below.

Table 25. Pearson Chi-square test results 

Chi-Square Tests
	Value	df	Asymptotic Significance (2-sided)	Exact Sig. (2-sided)	Exact Sig. (1-sided)
Pearson Chi-Square	5.111a	1	.024
Continuity Correctionb	4.382	1	.036
Likelihood Ratio	4.797	1	.029
Fisher's Exact Test				.029	.020
Linear-by-Linear Association	5.098	1	.024
N of Valid Cases	400
a. 0 cells (0.0%) have expected count less than 5. The minimum expected count is 14.25.
b. Computed only for a 2x2 table

From the test results, there is displayed a significant association between the relationship (marital status) of respondents and their respective levels of anxiety (χ2 (1) = 5.11, p = .02). This led to the rejection of the corresponding null hypothetical statement that the anxiety level of the respondents was independent of their respective relationship status; It also confirms further the suggested relationship from the contingency table. This leads to the conclusion that the anxiety levels of individuals under study was substantially impacted by relationship status, in other words, people of different relationship status recorded dissimilar levels of anxiety evidently.

This was tested using univariate analysis of variance test. This test was justified considering the fulfilment of variance equality assumption (Levene’s test (6, 394) = 1.34, p = .24). The results are presented in the table 26 below.

Table 26. Univariate ANOVA test results

Tests of Between-Subjects Effects
Dependent Variable:   Wellbeing
Source	Type III Sum of Squares	df	Mean Square	F	Sig.
Corrected Model	1015.521a	6	169.254	2.664	.015
Intercept	13765.201	1	13765.201	216.695	.000
Age	876.096	4	219.024	3.448	.009
RelationshipStatus	213.988	1	213.988	3.369	.067
Age * RelationshipStatus	379.387	1	379.387	5.972	.015
Error	24964.719	393	63.523
Total	61174.000	400
Corrected Total	25980.240	399
a. R Squared = .039 (Adjusted R Squared = .024)

From the table of results above, there is displayed a significant impact of age on the wellbeing of workers across the workaholic categories (F (4, 393) = 3.45, p = .01) thus leading to the rejection of the null hypothetical statement that the wellbeing of workers was independent of age category. In contrast, there was not established an evident association between the relationship status of the respondents and their respective levels of wellbeing (F (1, 393) = 3.37, p = .07) thus leading to then failure to reject the corresponding null hypothetical statement that the wellbeing of the respondents was not evidently influenced by their status of the relationship. Considering the objective of interest results, then there was established a significant interaction effect of age and relationship status of the respondents on the wellbeing of the respondents (F (1, 393) = 5.97, p = .02). This led to the rejection of the null hypothesis corresponding to the study objectives that there is no evident interaction effect of age and relationship status on the respondents’ wellbeing at work. Hence, it is concluded that the age of an individual based on whether the individual was married or not impacted significantly his/ her level of wellbeing.

The impact of age and relationship status on the perceived work environment of the bankers (participants)

This was tested using Univariate ANOVA test at 5% level of significance under the hypothesis that there is an evident impact of the relationship status and age on the perceived work environment of the bankers. This test was ascertained by the equality of variance assumption that was achieved from the Levene’s test (p > .05) and the results displayed below. 

Table 27. Univariate ANOVA results

Tests of Between-Subjects Effects
Dependent Variable:   PWE
Source	Type III Sum of Squares	df	Mean Square	F	Sig.
Corrected Model	2856.932a	6	476.155	.550	.770
Intercept	3516175.213	1	3516175.213	4062.675	.000
RelationshipStatus	63.807	1	63.807	.074	.786
Age	2428.125	4	607.031	.701	.591
RelationshipStatus * Age	145.717	1	145.717	.168	.682
Error	340134.708	393	865.483
Total	8633936.000	400
Corrected Total	342991.640	399
a. R Squared = .008 (Adjusted R Squared = -.007)

From the above table, age (F (4, 393) = .70, p = .59) and relationship status (F (1, 393) = .07, p = .79) were insignificant predictors of perceived wellbeing of bankers in Delhi as presented by the respondents. Similarly, there is no interaction between relationship status and age on the perceived wellbeing of workers (F (1, 393) = .17, p = .68). Hence, the above hypothetical statement is considered as false and a conclusion is made that the perceived work environment scores by bankers was not affected by their age or relationship status or the association between the two.

 Based on the above study objective results, it is observable that the anxiety level of the respondents is impacted by their respective age group. This is also true for their status of relationship where the married have a significantly different anxiety levels compared to the unmarried combining both the workaholic and the non-workaholic groups. In terms of wellbeing, then there is established a significant impact of age group on the wellbeing of respondents. This is however not true for the relationship status of work which indicates that both the married and unmarried had substantially equal levels of wellbeing. In addition, there is established an interactive impact of respondents age by relationship status on their wellbeing. In terms of perceived work environment, then there is no evident impact of age and relationship status on the perceived work environment same as the interaction between the two independent variables.

In this study, it was analyzed the anxiety symptoms, perceived work environment and the psychological wellbeing among then workaholic and non-workaholic groups among the banking employees in the Delhi. It involved individuals in active employment age, different job positions (trainee, mid-level managers, non-managers, top level managers and other managerial tasks), unmarried and unmarried relationship status, three education levels (bachelors, master degree and PhD), males and females, workaholic and non-workaholic states, and moderate and low anxiety levels. While the respondents had a normally distributed perceived work environment scores, most of the respondents recorded lower scores of anxiety.

Analyzing the study regarding anxiety symptoms, perceived work environment and psychological wellbeing of bank employees in Delhi was done using six major objectives. In the first objective, it was sought to establish the significance in the difference between workaholic and non-workaholic group over anxiety, psychological wellbeing and the perceived work environment. This was answered in three sections. In the first one, the relationship between the workaholic and anxiety level conditions was tested; this was discovered to be significant in which case, those who had low anxiety levels were mainly the non-workaholic category whereas the workaholic groups mainly experienced moderate level of anxiety. Secondly, it was sought to discover the association between the workaholic group and the psychological wellbeing of the respondents. In this case, a significant association was established thus indicating that the workaholic category had a substantially different level of wellbeing relative to the non-workaholics. Finally, the perceived work environment with the workaholic state of individuals was tested; this resulted in an insignificant association between the variables showing that workaholic and non-workaholic groups had the same level of perceived work environment scores. Therefore, it was concluded from the first objective that the workaholic state of the Bank employees in Delhi significantly influenced their anxiety levels as well as their extent of wellbeing. Nevertheless, the employees’ state of workaholic had no evident impact on the extent to which they mastered their work environment.

The second objective sought to establish the association between gender and the workaholic state of the Delhi bankers. In this state, there was no evident association between gender and the workaholic state of the respondents. This implies that the workaholic state of the Delhi bankers was independent of whether the respondents were males or females. The third objective was on the gender difference on the workaholic and non-workaholic groups over anxiety levels, psychological well-being, and the perceived work environment; this was answered in three parts. The first part was on the association between anxiety levels and gender across the workaholic groups where gender and the workaholic state of the respondents had no evident impact on their level of anxiety. The second part was on the association between gender and wellbeing being among the workaholic and non-workaholic groups. As earlier established, the wellbeing of individuals was evidently impacted by the workaholic state of individuals. Nevertheless, gender had no evident impact on the wellbeing of the Delhi bankers. Finally, there was no interaction effect of gender and the workaholic state of individuals on their level of wellbeing among the Delhi bankers. This implies that the wellbeing of male and female bankers under study was not impacted by whether they were workaholic or non-workaholic. Hence, it was concluded that though the wellbeing of the Delhi bankers was evidently impacted by their workaholic category, the workaholic nature did not influence the level of wellbeing between male and female bankers. In the third part of the objective, the gender difference was tested on the workaholic and non-workaholic groups’ scores of perceived work environment. In this category, gender had no evident impact on the perceived work environment score same as the interaction between workaholic state and gender. This implies that the perceived work environment scores by the selected respondents was influenced by other factors rather than the gender and workaholic state; equally, the perceived work environment scores between males and females was independent of whether they were workaholic or not. Combining the three sets of objectives, then it is clear that while the anxiety levels and wellbeing of workers as well as their scores of the work environment was not impacted by gender and the interaction between gender and their workaholic state. Hence, other factors should be considered rather than these two.

In the fourth objective, the impact of perceived work environment on anxiety and psychological well-being between the workaholic and non-workaholic bankers in Delhi was tested under three categories. In the first step, the psychological wellbeing and its relationship with the perceived work environmental scores was tested among the workaholic group was tested; this resulted in an insignificant negative association. The second test was performed among the non-workaholic categories; this also resulted in an insignificant negative relationship between wellbeing and perceived work environment scores. Hence, it can be concluded that there was no evident association between the wellbeing of the Delhi bankers and their respective knowledge of the perceived work environment. Hence, other factors rather than perceived work environment scores should be considered. Thirdly, the impact of perceived work environment on the worker anxiety levels was tested between the workaholic and non-workaholic groups. In this section, though there was no evident impact of perceived work environment scores on the anxiety levels of the respondents, there was an interaction between the variable and workaholic state of the respondents on their levels of anxiety. Hence, it implies that the anxiety level of the bankers was affected by their perceived work environment depending on whether they were workaholic or non-workaholic. The general conclusion based on the fourth objective that the perceived work environment scores of the Delhi bankers had no impact on their corresponding levels of anxiety and wellbeing across the workaholic and non-workaholic categories; this was specific to individual variables. Nevertheless, the anxiety level of the bankers was impacted by the perceived work environment scores based on their workaholic state.

The second last objective was to establish the impact of job position on the anxiety levels, the psychological wellbeing of the participants and the perceived work environment between the workaholic and non-workaholic groups. This objective was answered in three categories. In the first category, it was sought to discover the association between job position and the anxiety condition among the workaholic and non-workaholic groups. There was not discovered any form of relationship between the job position and the anxiety level of the bankers in the workaholic group considering that this category only had moderate level anxiety. In contrast, the non-workaholic group recorded an insignificant impact of job position on the level of anxiety among the non-workaholic group. In total, there was established a significant association between the job position of bankers and their respective levels of anxiety combining the workaholic and non-workaholic groups. Hence, it can be concluded from this sub category that though the job position of workers had a significant impact on the anxiety levels of workers, this did not vary between the workaholic and non-workaholic categories as evident by the non-significant level or non-existent associations recorded. Secondly, it was sought to establish the impact of job position on the Delhi banker’s wellbeing between the workaholic and non-workaholic individuals. In this category, there was discovered a significant impact of workaholic state on the respondents’ wellbeing. Nevertheless, the job position of individuals did not display a significant impact on the wellbeing of the bankers. This was the same result for the association between the wellbeing of the respondents and the interaction effect between their workaholic state and the job position. Finally, it was sought to discover the impact of job position on the perceived work environment of the participants; it was established from this study the non-existence of a significant impact of the variable on bankers’ perceived wellbeing. It is thus concluded that the influence of worker wellbeing and perceived work environment by the job position of bankers depending on whether they are workaholic or not is insignificant and thus other factors should be considered. On the other hand, the anxiety levels of workers is affected by their job position.

Finally, it was sought to establish the impact of age and relationship on the anxiety levels, worker wellbeing, and perceived work environment across the workaholic and non-workaholic categories. This was answered in four categories. The first one was on the impact of age on the anxiety levels between the workaholic and non-workaholic categories. This was established a significant impact of the age on the anxiety levels of the Delhi bankers under study. Secondly, it was sought to establish the association between relationship status and anxiety levels between the workaholic and the non-workaholic categories. Like the first sub category, there was displayed a significant impact of relationship status and the levels of anxiety of the respondents. In the third category, the objective was to establish the impact of age and relationship status on the wellbeing of the bankers. The first result was a significant association between age and the wellbeing of the study participants; this was however not true for the relationship status which displayed an insignificant impact on the worker wellbeing. In addition, there was discovered a significant interaction impact of age and the participants’ relationship status on their wellbeing. Finally, it was sought to establish the influence of age and relationship status on the perceived wellbeing of Delhi bankers in which case, there was not established an evident influence of age and relationship status, both individually and through their interaction, on the perceived work environment scores by workers. Hence, it is concluded from the general objective that the level of wellbeing and anxiety level of workers varied across ages and relationship status. As well, the impact of relationship status or age on the wellbeing scores of the Delhi bankers was significantly impacted by their interaction with each other. In other words, the age influence on the participant wellbeing was influenced by whether the respondent was married or unmarried. In contrast, the perceived work environment by workers was not affected by the age and relationship status of the respondents.

In this study, it is discussed the study results in the context of past works of literature, in other words, comparing the study to past study results with an aim of achieving generalizability. In the study, it was established from the first objective that the workaholic state of an individual had a substantial impact on their level of anxiety and psychological wellbeing but not their perceived work environment scores. This is justifiable considering that by definition of the workaholics, individuals that are not capable of limiting the time they spent at work irrespective of the negative results experienced including health or relationship damage. In particular, the workaholics, as stated by Hussein et al. (2016) results in life imbalance that directly contributes to stress and depression, a result similar to the one of this study. Equally, they established that other than stress, being a workaholic results in a life imbalance that largely result in the deterioration of mental and physical health due to fatigue and depression which directly causes hence linking the working condition to stress levels. In a different study by Gabriele et al. (2017) regarding the work-related stress in the banking sector, revealed that bankers who spent much of their time at work and without proper balance ended up more stressed relative to those who had their time balance. On top of the time imbalance stress, the condition leading to workaholic state, according to Emily et al. (2020) also contributed largely to the stress by workers; an instance in this situation is the need for more money to cater for their needs which make an individual work extra times without having balance for other activities in life. Other than stress, workaholic state of individuals, as stated by Gayle (2001) in his paper about the workaholic tendencies and the high potential for stress among co-workers discovered that this condition extends to the high levels of depression among workers, increased anxiety levels, and high level mood disorder. This was a similar finding to that of Shahnaz and Jamie (2008) who discovered that due to the workaholic state of the workers including bankers, their overall wellbeing is negatively affected. In a research by Octavian and Nicoleta (2020), those who spend a lot of their time at work without making a balance to other needs in life end up not making a strong social relationship with their friends and families; this usually leads to seclusion or sometimes divorce between partners, a factor that mainly reduces the level of wellbeing among the employees. Finally, workaholism, as stated by Sania and Saadia (2019) in their paper about workaholism being a predictor of work-family conflict, established that the condition largely increases the work-family conflict and hence the overall wellbeing of the worker which is the center of conflict; this combines the private and public sectors. Therefore, it can be concluded that the workaholic category of employees including bankers significantly experiences low level of wellbeing and high level anxiety as established in the results of this study. In terms of perceived work environment, there are two possible impacts of workaholism on the perceived work environment by the bankers. First, it can lead to the improved or reduced scores of the perceived work environment and second, it cannot have an impact depending on the work condition. For instance, take a situation where a workaholic banker is feeling good at job in terms of environment and treatment from colleagues and management, then such an individual is likely to score high in terms of positive perceived work environment thus explaining the first possibility. In the second possibility, assuming one is compelled by other factors such as home demands is likely to be a workaholic despite the scores in the perceived work environment thus suggesting the second possibility of non-relationship. In justification to this, a study by Giovanni and Maria (2018) in their aim to establish the differential effects of workaholism and work engagement on the life and work domain interference discovered that there is no direct impact of being a workaholic on the perceived work environment. Instead, there are other interactive factors such as place of work of an individual, role and the interest of a person on the level of knowledge by a worker regarding the place of work. In the general view of the first objective, it is thus concluded that the workaholic state of the Delhi bankers significantly impacted their anxiety and psychological wellbeing but not the perceived work environment.

In the second objective, the relationship between gender and the workaholic state of an individual was established as insignificant. This is justifiable considering that workaholism is strongly driven by other factors other than gender. For instance, a person maybe workaholic due to the burdens and responsibility and hence the need for workaholism-accompanying benefits like overtime allowance. Equally, the desire to stay longer at work can also be driven by those who want spend less time at home with their families due to social factors or even less time performing other businesses. In line with this objective, a similar finding was established by Iwona and Malwina (2019), who while studying gender difference in workaholism and work related variables, discovered that the workaholic nature of an individual was mainly driven by needs and nature rather than the gender of a person. Instead, Shahnaz and Jamie (2008) established that gender mainly influenced the level of impacts of workaholism. Specifically, they established a significant difference in the impact of workaholism on anxiety and the overall wellbeing of workers where females were more likely affected than males. In a different study by Sarfaraz et al. (2021) on the impact of workaholism and work engagement on the employees’ stress and satisfaction, there was discovered a significantly different levels of workaholism and work engagement on the stress levels and the overall employee satisfaction across gender. In particular, males were less affected than females considering the difference in their psychological setup. It is therefore concluded that though there is not a significant impact of employees, including Delhi bankers, gender on their workaholic states, the workaholism effect on the employees varied substantially across gender.

For the third objective, there was not established a significant impact of gender difference between the workaholic and non-workaholic group on the overall wellbeing and the perceived work environment of the Delhi banker’s respondents. The perceived work environment result was justifiable considering that how one perceives the environment at work, as earlier stated, determined by factors as discrimination and kind treatment at work rather than gender or workaholism or the interaction between the two. Nevertheless, the wellbeing level in correspondence to gender, workaholism and the interaction between the two varies depending on the dimensions. For instance, in a situation where there is a little extent of workaholism impact on workers, then there is a likelihood that such an impact will not be different across gender. In contrast, assuming a situation where there is a strong impact of workaholism on the workers in regards to wellbeing, then it is more likely that gender will have an impact since the psychological factors differs substantially across gender. By comparison to past works of literature, then this study results regarding the perceived work environment score is similar to that of Dina et al. (2012) on workaholism and self-efficacy as initiators of the job demands-resources model who discovered that interest is a great driver on the perceived work environment alongside the level of orientation compared to gender or workaholism or the interaction between the two variables. Similarly, Cecilie et al. (2014) on their study of workaholism prevalence, found that gender had no evident influence on the wellbeing of workers solely or through its interaction with the workaholic variables. An example in this situation according to Iwona and Malwina (2019) is where males and females have an almost equal family responsibility such as taking care of ailing ones, then it becomes an equal content of stress between them. In contrast, a finding by Diana and Aleksandra (2021) on the workaholism components in relation to work and life values was that workaholism had an impact on the wellbeing of workers, a finding that substantially differs between males and females. In particular, they discovered that gender considered more vulnerable than the others like the females and other gender categories compared to males are more likely to experience the wellbeing problems. By providing example, Dina et al. (2012) stated that a woman with parental role over kids and is workaholic finds it difficult to meet the two responsibility levels compared to a male without such responsibility. Similarly, a bisexual or a homosexual individual who has the fear of rejection and possible attack from individuals in the society, in a workaholic state, may extend time at work forcefully to avoid social judgement which may extend to physical violence; this contributes to the negative wellbeing of such a worker. As such, gender and the workaholic state is observed to significantly impact the level of wellbeing of the workers. Hence, it can be concluded from this objective that the impact of workaholism between males and females have no substantial impact on the perceived work environment by an employee. Nevertheless, the above factors have a divergent impact on the worker wellbeing depending on the research setup.

In the fourth objective, the impact of perceived work environment scores on the psychological wellbeing was discovered as insignificant between the workaholic and non-workaholic categories. This is attributable to the fact that at places of work, factors such as discrimination, workload extent and the environmental condition of workplace are substantial influencers of the psychological wellbeing of workers rather than the one’s perception of the work environment. This, according to Satoshi and colleagues (2022) is true since the perception of the work environment is either positive, neutral or negative thus have a varying impact on the level the levels of anxiety and wellbeing where neutral grounds implies no effect. In a different objective subset, there was established an interactive impact of perceived work environment and the workaholism condition of the respondents on their wellbeing. In this case, it can be justified that the workaholic state of an individual being a significant determinant of worker wellbeing, this level of impact can vary based on the level of perceived work environment. In line with this, Dina et al. (2012) stated that the workaholic workers are in most instance sensitive to at work treatment, a factor that affects their anxiety levels and hence the general level of psychological wellbeing. This was a similar finding to Sania and Saadia (2019) who stated that though the perceived work environment alone cannot affect the stress levels of a bank employee, it displays an evident impact when interacted with other factors such as leadership roles and the workaholic state of the corresponding workers. Therefore, it can be concluded based on this objective that through this study results and those of past works of literature, there is no evident association between the perceived work environment and the workers’ wellbeing, this variable effect on the worker wellbeing differs significantly between workaholics and the non-workaholics.

In the fifth objective, in the association between job position and anxiety, the psychological wellbeing, and perceived work environment of Delhi, it was discovered a significant impact of job position on the anxiety levels of workers between the workaholic and the non-workaholic. Nevertheless, the job positon of workers had no evident influence on their wellbeing as well as the interaction between the two workaholism and job position; this was a similar case to the job position impact on the perceived work environment. The result on the level of anxiety is justifiable considering that the work level differs from one job position to another including administrative roles. For instance, while managers are concerned with receiving reports and presenting them to the board of directors as well as controlling the general operation of the business, junior employees, though without administrative responsibilities, they receive more pressure to deliver thus are likely to have more pressure which directly affects their anxiety levels. In an explanation to this, Concetta and Marco (2020) discovered that the employees at lower levels are mostly subjected to pressure of work delivery, a factor that contributes to them being more stressed and anxious. Equally, this group suffers from lower pay and financial benefits, a factor that may subject them to mental stress and hence the declining mental health. The results on the wellbeing by the interaction between the workaholism categories and job position, this was different from previous study by Thomas et al. (2020) who discovered that the wellbeing of workers may varies across job positions under the significant influence of the workaholic state. This displays a need for a continuous study to establish exactly the influence of the job position due workaholic state of individuals on their overall wellbeing. While explaining the perceived work environment relationship with job position, Concetta and Marco (2020), stated that though the job position matters, it does not have a lone impact unless moderators such as benefits arising from position and discrimination in promotion are involved. It can, hence, be concluded that while the anxiety levels of the Delhi banker’s wellbeing are attributable to the job position, this factor have no evident impact on their wellbeing and perceived work environment among the workaholic and non-workaholic categories.

In the final objective, the test on the age category and relationship status on the anxiety levels, psychological wellbeing and perceived wellbeing of the Delhi bankers established a significant impact of age category and the state of relationship on the anxiety levels. Equally, the age category displayed a significant impact on the bankers’ wellbeing whereas the relationship status did not. In terms of interaction, the influence of age group on the wellbeing of bankers’ varied between the married and unmarried. In contrast, age and relationship status had no influence on the perceived wellbeing of the Delhi bankers. Justifying this study results, then age comes with responsibility and experience required to manage the levels of anxiety and the psychological wellbeing among workers. Similarly, the state of relationship defines the responsibility by a person, for instance, a married person has to take care of more factors and this makes such an individual to be more anxious at work on how to earn more. As well, age having proven to evidently impact wellbeing is influenced by the marital status which comes with new social responsibilities and psychological impacts. Checking into past works of literature discussing the same, then the study results on age, relationship status and the level of anxiety is similar to that of Xin et al. (2022) who stated that in most cases, age and relationship status comes with more responsibility and urge to meet the rising demand. As a result, the mental wellbeing and the level of anxiety are affected in this case. In addition, they state that the relationship impact can exceed the control extent on the level of wellbeing, a factor that sees a stronger level of interaction impact between age and relationship status on the worker wellbeing. For the perceived work environment, then this solely depends on at work treatment rather than the marital status and age. Therefore, it is concluded that while considering the different factors affecting bankers’ wellbeing and anxiety levels, it is imperative to also include age group and state of relationship. On the other hand, the two factors should not be included while addressing perceived work environment.

In general, it is evident that the study results were mostly similar to those of past works of literature. This promoted the level of generalizability and hence the attainment of external validity. However, it also pauses a chance for future study to establish a harmonizing result based on the results that did not match past works of literature.

While the study achieved the desired objectives, the study had three major limitations. The first limitation of the study was on the sample size. This is considering that the bankers in new Delhi are many and thus need a larger representation of between 1000 to 1200. Secondly, there is experienced error of omission in selecting the factors affecting the psychological wellbeing and anxiety levels among the workaholic and non-workaholic bankers. This likely resulted in confounding effects more so for the univariate ANOVA analysis thus affecting the study effect size. Finally, the study was subject to bias in the selection of the workaholic and non-workaholic individuals where majority were non-workaholic. This likely affected the overall study significance. It is therefore important that in the future research, these factors including increased sample size, increased number of influencers to anxiety levels and the overall banker wellbeing and the balancing of workaholic and non-workaholic categories be considered to ensure the achievement of better study results. 

Based on the study results, the following recommendations are made. First, while addressing the anxiety condition, the psychological wellbeing of workers and their respective perceived work environment, it is relative that the workaholic state be taken into consideration being that it has a substantial impact. Secondly, job position should be considered while analyzing the anxiety levels of the Delhi bankers but not when handling perceived work environment. Thirdly, the job position and relationship state of the bankers should be highly considered while addressing the anxiety and psychological wellbeing of the Delhi bankers but not perceived work environment. This should also include the interaction effect between the age category of bankers and their respective relationship status. Finally, while addressing psychological wellbeing and anxiety levels among Delhi bankers, the interaction between workaholism and perceived wellbeing of workers should be considered.

Conclusion

Based on the study results, it is concluded that while addressing matters concerning Delhi bankers’ state of wellbeing, workaholic state of individuals plays a significant impactor as well as the age group, job position for anxiety levels, and state of relationship. This study results can be used alongside other works of literature to arrive at a conclusion regarding the influential factors affecting the Delhi bankers’ wellbeing and the anxiety levels. Equally, it can be used as a background for future scientific research that will define exhaustively the impactors of Delhi bankers’ anxiety levels and employee wellbeing. Finally, the study results can be used to for the purposes of learning by other individuals working on similar projects.

Having earlier discussed the study limitations, I suggest a further study that will take into consideration a larger sample size with additional prospected influencers of anxiety and wellbeing among the Delhi bankers, and a well-balanced workaholic vs non-workaholic categories. In this study, I suggest the use of a multiple linear regression that will establish individual and joint influence of variables on the bankers’ wellbeing as well as the extent of influence. From the suggested study, I suggest a comparison to this study to authenticate the study results in regards to factors affecting the anxiety levels and wellbeing of Delhi bankers.

References

Cecilie, S., Mark, D., Jorn, H., Fredrik, J., & Stale, P. (2014). The Prevalence of Workaholism: A Survey Study in a Nationally Representative Sample of Norwegian Employees. PloS ONE, 9(8).

Concetta, R., & Marco, T. (2020). Mental Well-being Among Workers: A Cross-national Analysis of Job Insecurity Impact on the Workforce. Social Indicators Research, 152, 421-442.

Diana, M., & Aleksandra, T. (2021). Workaholism Components in Relation to Life and Work Values. International Journal of Mental health and addiction, 19, 529-545.

Dina, G., Silvia, S., Wilmar, B., & Marco, D. (2012). Self?efficacy and workaholism as initiators of the job demands?resources model. Career Development International, 17(4), 375-389.

Emily, M., Shahnaz, A., Wuensch, K., & Dolbier, C. (2021). Work hard, play hard…or maybe not: A look at the relationships between workaholism, work-leisure conflict, and work stress. Journal of Leisure Research, 52(3).

Gabriele, G., Arcangeli, G., Perminiene, M., Lorini, C., Ariza-Montes, A., Javier, F.-P., . . . Mucci, N. (2017). Work-Related Stress in the Banking Sector: A Review of Incidence, Correlated Factors, and Major Consequences. Frontiers in Psychology.

Gayle, P. (2001). Workaholic Tendencies and the High Potential for Stress Among Co-Workers. International Journal of Stress Management, 8, 147-164.

Hossein, J., Nejad, B., Abadi, F., Haghi, R., & Hojatinasab, M. (2016). Relationship of Workaholism with Stress and Job Burnout of Elementary School Teachers. Health, 8(1), 1-8.

Iwona, D., & Malwina, S. (2019). Gender differences in workaholism and work-related variables. Studia humanistyczne agh, 18.

Octavian, D., & Nicoleta, D. (2020). Workaholism in IT: An Analysis of the In?uence Factors. Administrative Sciences, 10(96).

Sania, T., & Saadia, A. (2019). Workaholism as Predictor of Work-Family Conflict and Mental Well-Being: Comparison of Public and Private Sector Employees. Pakistan Journal of Commerce and Social Sciences, 13(2), 419-435.

Satoshi, A., Fumiaki, K., & Shobei, Y. (2022). The Antecedents and Consequences of Workaholism: Findings From the Modern Japanese Labor Market. Frontiers in Psychology.

Shahnaz, A., & Cunningham, J. (2008). "Workaholism, work stress, work?life imbalance: exploring gender's role". Gender in Management, 23(8), 553-566.

Thomas, C., Line, R., Malene, F., Torres, T., Ida, & E. (2020). Job autonomy and psychological well-being: A linear or a non-linear association? European Journal of Work and Organizational Psychology, 395-405.

Xin, C., Heng, Z., Peng, L., & Xiaozhi, H. (2022). The Influence of Mental Health on Job Satisfaction: Mediating Effect of Psychological Capital and Social Capital. Frontiers in Public Health.