The effect of employment status upon stress and burnout levels

The effect of employment status upon stress

and burnout levels

by Shamel Rajapakse

Abstract

A data collection and analysis activity was carried out by a class of university students in order

to complete course requirements of the module research methods. As required by the study

students recruited and analyzed data regarding participants through the questionnaire method of

survey. Administration of questionnaires and data collection was supervised by the instructor of

the module. Based on data collected through this study, this analysis examines how stress level

differs between employed and unemployed individuals. Employment categories taken under

consideration as employment are full time employed, part time employed, self-employed and

student. Retired and unemployed categories are collectively considered as unemployment. A

total of 99 participants were recruited for this study. A demographics details questionnaire, a

stress and burnout questionnaire and a burnout prevention questionnaire were administered to

each participant. The results revealed an increase in stress and burnout scores in the employed

category of participants when compared to the unemployed category. However, effect size

between the two variables was poor.

Key words: stress, burnout, employment status, unemployment, student stress, work load, job-

stress

Introduction

In an overly competitive and time-driven world, terms such as “stress” and “burnout” are not

uncommon. Stress is understood as having a negative psychological impact and even able to

physical diseases such as cardiovascular diseases and cancer (Cohen, Janicki-Deverts and Miller

2007). Multiple studies have been conducted in recent research in order to establish links

between stress levels and employment related situations (Tennant 2001, Shields 2006, Prottas

and Thompson 2006, Anisman et al. 2005). Implications of employment statuses such as

unemployment, full time employment, self-employment, studentship have been taken under

study regarding work load and one’s ability to cope with external pressures (Tennant 2001,

Leiter, Maslach and Schaufeli 2001, Castillo and Misra 2004, Bartley 1994, Kim and Moen

2002). Psychological stress is defined by Cohen, Gordon and Kessler as a condition that occurs

“when an individual perceives that environmental demands tax or exceed his or her adaptive

capacity” (1995). Leiter, Maslach and Schaufeli define burnout as “a prolonged response to

chronic emotional and interpersonal stressors on the job, and is defined by the three dimensions

of exhaustion, cynicism, and inefficacy” (2001). As burnout is considered to be a result of

stressors experienced during employment, it is intriguing to measure how stress levels vary from

employed individuals to unemployed. Furthermore, organizational and workplace demands are

viewed as a common cause of stress in an individual’s life, relatively matched only by demands

in the family domain (Tennant 2001). Stress and burnout caused by employment leading to job

burnout can be influenced by factors such as personality traits, organizational settings and

demographic conditions (Leiter, Maslach and Schaufeli 2001). Academic demands have also

been found to cause stress in the student population, as revealed by studies conducted on college

students (Castillo and Misra 2004). According to research conducted by Castillo and Misra,

“psychological, emotional, behavioral and cognitive” are categories of stressors that affect the

psychological wellbeing of college students (2004). In addition to stress experienced in

employed conditions such as work and academics, it is important to note that research has found

evidence to show the possibility of psychological stress in unemployed and retired individuals as

well (Kim and Moen 2002, Creed and Macintyre 2001). Therefore it is clear that psychological

stress and burnout are conditions that apply to a wide range of employment situations. However,

a comparative study between the differences of stress and burnout levels in unemployed and

employed situations has not been done by any of the above mentioned studies. Furthermore, it is

important to note that the above studies have been conducted in the Western world, and therefore

remain questionable regarding their relevance to Asian cultures such as Sri Lanka. In this present

study conducted in Sri Lanka, 99 participants are recruited to be measured regarding

demographic details and stress and burnout. The relationship between employment,

unemployment and stress levels is taken under study. The research hypothesis (H1) of this study

is that employment (inclusive of full time employment, part time employment, self employment

and studentship) causes significantly increased stress in individuals when compared to

unemployed and retired individuals’ stress levels. The null hypothesis (H2) of this study is that

employment status has no significant effect upon stress levels of individuals.

Methodology

Survey design

The method of questionnaire in survey was selected for this study. The study was designed as a

combined effort by the students who were required to conduct the study and the module

instructor. A demographics questionnaire was designed by the students which included 14 close

ended multiple-choice categorical questions and 1 open ended question. The demographical

categories assessed by the questionnaire consisted of gender, age, highest level of education

completed, current level of employment, current income level, number of hours spent working

(in job) or in formal education (lectures) per week, relationship status, most frequent mode of

transport, average number of hours of sleep per day, ethnicity, religion, reported life satisfaction,

reported stress level and perceived physical appearance. The open ended question inquired about

three changes, if any, the participant desired to implement in life. The Stress and Burnout

Questionnaire and Burnout Prevention Assessment were provided by the instructor. The stress

and burnout questionnaire contained 24 statements that required a score of 0, 1 or 2 to be given

according to applicability to participant in the past 3-6 months. The 25th statement contained

somatic symptoms of stress which required the participants to provide the sum of symptoms

experienced frequently. The total score achievable from this questionnaire amounted to 50, and

each participant’s score was recorded at the close of the questionnaire by the researchers. The

burnout prevention assessment included 23 questions measuring participant responses in a range

of 0-5, 0-3 or 0-4 according to a range of categorical responses. The responses were aimed at

measuring how often the participant experienced symptoms of stress and burnout. The total score

of the burnout prevention assessment was 100. The score achieved by each participant was

recorded at the end of the questionnaire by the researchers. A score of over 60 implied that ample

measures are in place to prevent burnout, while a score of over 40 indicated sufficient measures

to be in place as well as the need to secure additional methods of burnout prevention. A score

under 30 indicated the need to prioritize implementation of methods in order to prevent burnout.

Data collected through the questionnaires was entered into a single worksheet and analyzed as a

whole by the use of descriptive statistical measures. Each student researcher was then allowed to

select a research hypothesis and conduct a research paper upon the selected hypothesis. The mid

value for stress and burnout is considered as 25, as the total score amounts to 50. The mid value

for burnout prevention is considered as 50, as the total score is 100.

Participants

Participant recruitment was conducted through students. Each student recruited three participants

for the study. Participants consent was acquired through informative consent forms that

explained the nature and proceedings of the study. Participants were required to be present at a

given location in the university premises where the study was conducted simultaneously to all

participants by student researchers. The questionnaires were administered to participants by the

students under the supervision of the module instructor. The complete number of participants in

the study amounted to 99. Participants were aged between 18 years to 55 years and above, and

included 53 females and 46 males. Accordingly the sample consisted of 46.50% male

participants and 53.60% female participants. Educational levels of participants varied from

Ordinary Level Examinations or equivalent to post graduate level. Participants’ income levels

varied from below LKR 15,000 per month to above LKR 100,000 per month. An overview of

demographic details of the participants in provided in the results section (figure 1).

Results

The demographic details questionnaire revealed the following general information regarding the

demographic representations of the sample (figure 1).

Figure 1. Demographic details of participants

Demographical categoryNumber of participants and

percentage

Gender

Male

Female

46 (46.50%)

53 (53.40%)

Age

18-24 yrs

25-34 yrs

64 (64.65%)

13 (13.10%)

35-44 yrs

45-54 yrs

55 and above

4 (4.05%)

8 (8.09%)

10 (10.10%)

Highest level of education

O/L or equivalent

A/L or equivalent

Diploma or professional training

University degree

Postgraduate qualification

10 (10.10%)

45 (45.50%)

23 (23.23%)

17 (17.17%)

4 (4.05%)

Current employment status

Employed full time

Self employed/working part time

Neither student nor employed

Student

Retired

20 (20.20%)

18 (18.20%)

15 (15.20%)

47 (47.18%)

4 (4.05%)

Current income level

Below LKR 15,000

Between LKR 15,000-30,000

Between LKR 30,000-50,000

Between LKR 50,000-100,000

Above LKR 100,000

*Participants who did not respond

29 (29.30%)

9 (9.09%)

14 (14.15%)

7 (7.08%)

8 (8.09%)

31 (31.31%)

Number of hours spent working per week

Less than 10 hours

Between 10-20 hours

Between 20-30 hours

Between 30-40 hours

Between 40-50 hours

More than 50 hours

*Participants who did not respond

19 (19.20%)

24 (24.26%)

13 (13.13%)

13 (13.13%)

16 (16.17%)

7 (7.08%)

7 (7.08%)

Relationship status

Single

In a relationship

Married/Engaged

Divorced

Other

46 (46.50%)

20 (20.20%)

30 (30.30%)

1 (1.01%)

2 (2.02%)

Most frequent mode of transport

Private transport, but not self-driven

Private transport, self-driven

Motorbike

Taxi/Three-wheeler/Cab

Public transport

Bicycle

On foot

13 (13.13%)

14 (14.15%)

7 (7.08%)

33 (33.33%)

30 (30.30%)

0 (0%)

0 (0%)

Average amount of sleep per day

Less than 6 hours

6-8 hours

8-10 hours

More than 10 hours

25 (25.26%)

65 (65.66%)

9 (9.10%)

0 (0%)

Ethnicity

Burgher

Muslim

Sinhalese

Tamil

Other

3 (3.03%)

42 (42.42%)

40 (40.50%)

12 (12.12%)

2 (2.02%)

Religion

Buddhism

Christianity

Hinduism

Islamism

Atheism

Do not identify with any religion

29 (29.30%)

17 (17.18%)

8 (8.09%)

42 (42.40%)

1 (1.01%)

2 (2.02%)

Reported life satisfaction

Somewhat satisfied

Very satisfied

Somewhat dissatisfied

Very dissatisfied

*Participants who did not respond

39 (39. 40%)

53 (53.50%)

4 (4.05%)

0 (0%)

3 (3.03%)

Reported stress

Almost daily

Several times a week

About once a week

Once or twice a month

Only when faced with a new or demanding

situation

Rarely

Almost never

7 (7.08%)

15 (15.20%)

11 (11.00%)

6 (6.10%)

42 (42.50%)

12 (12.00%)

2 (2.00%)

Perceived physical appearance

Very happy with physical appearance

Overall happy, would change few minor

things

Want to change several things about physical

appearance

Extremely dissatisfied with physical

appearance

38

56

4

1

In relation to employment statuses of the sample, 20 participants (20.2%) of the sample belonged

to the full time employed category. 18 participants (18.2%) belonged to the self-employed/part

time employed category. 15 participants (15.16%) belonged to the unemployed category. 47

participants (47.48%) were students. 4 participants (4.05%) were retired. The average burnout

prevention score for full time employed participants was M=57.5. Self employed participants

received an average score of M=17.17. Average stress and burnout score of unemployed

participants was M=17.13. Stress and burnout average of students amounted to M=17.02.

Average stress and burnout score of retired participants was M=9.25. In the calculation of

statistical significance, employment status is regarded as the independent variable and stress and

burnout score is the dependable variable.

The highest stress and burnout score was found in the self employed/part time employed

category. The second highest stress and burnout score belonged to full time employed

participants. Students, retired participants and unemployed participants held third fourth and fifth

ranking stress and burnout scores. An overview of stress levels related to each employment

status is displayed in the graph below (figure 2).

Employed

Full t

ime

Self e

mployed/e

mploye

d part tim

e

Unemplo

yed

Studen

t

Retired

050

100150200250300350

Figure 2. Collective stress and burnout scores of each em-ployment category

Employment status

Stre

ss a

nd b

urno

ut sc

ores

Accordingly, the collective score of the stress and burnout questionnaire for full time employed

participants was 290. Self employed/part time employed participants scored 309 collectively.

Unemployed participants scored 15. Students scored 257, and retired participants scored 37.

Therefore it is evident that full time employment, part time employment/self employment and

studentship caused higher test scores in the sample when compared with stress and burnout

scores caused by unemployment and retirement. The range of variability among stress and

burnout scores of each employment status was: full time employed participants 32, self

employed/working part time 27, unemployed 23, student 34 and retired 10.

In order to analyze the effect of employment status upon stress levels, the data is divided into

two main categories of employed and unemployed. Employed scores include full time employed,

self-employed/part time employed and student categories of demographic data. Unemployed

includes unemployed and retired categories of demographic data. 80 participants belonged to the

employed category, while 19 belonged to the unemployed category. The mean of the stress and

burnout scores of employed category was M=16.5. In the unemployed category, M=15.5. The

standard deviation of the collection of stress and burnout scores on the employed category

(including full time employed, self employed, part-time employed and student categories) was

ơ=8.28. The standard deviation of the scores of the unemployed category (including retired) was

ơ=6.96. The standard deviation value for the overall score of both employed and unemployed

categories was ơ=8.01. A comparison of stress and burnout scores of the above mentioned two

categories is summarized in the graph below (figure 3).

Employed Unemployed0

200

400

600

800

1000

1200

1400

Figure 3. comparison of stress and burnout scores between employed and unemployed categories

employment category

stre

ss a

nd b

urno

ut sc

ores

The effect size of the difference between stress and burnout scores of the two categories was

computed to be d=0.1312. As effect size was below 0.3, the effect size is considered poor.

Discussion

Results indicated that a relationship does exist between employment status and stress and

burnout scores. Stress and burnout scores of full time employed, self employed/part time

employed and student categories yielded higher than stress and burnout scores of unemployed

and retired categories. However, as the effect size of the result was poor, the generalizability of

the results to a larger population remains questionable. It must also be noted that the student

category constructed the majority of the sample (47.47%). It is questionable as to whether

studentship can be considered as a valid form of employment, as the pressures concerning work

stress and job burnout may not be equally applicable to students. Therefore the inclusion of

students in the employed category creates complications in the interpretation of the results.

Similarly, only four participants of the sample belonged to the retired category. Therefore the

stress and burnout scores of the retired category may have faced a statistical disadvantage, as an

increased number of retired participants may have altered the results of the study. The

categorization of unemployed individuals and retired individuals into one category as

unemployed may also be inaccurate, as stress levels of an unemployed individual may be

significantly different to stress levels of a retired individual (Bartley 1994, Kim and Moen 2002).

The participants of the research were selected through random sampling and therefore favor the

generalizability of the results. A near equal representation of genders is also evident in the

sample (46.50% male participants and 53.40% female participants). However, the participants of

the research were recruited by students from an urban university, and therefore belonged to an

urbanized culture. Therefore the extraneous variable of urban effect upon stress may have

influenced the results of the study (Fiorito et al. 1991, Deuschle et al. 2011). The sample size of

99 may also have been inadequate to measure the difference in stress levels of employed and

unemployed categories, and reduced variation within the sample, such as too few retired

participants and too many student participants. Therefore increased variety and quantity in the

research sample may benefit future replications of this study.

Several flaws and limitations were also found in the survey design. The demographic details

questionnaire contained several double barrel response categories that may have confused the

participant in answering and the researcher in interpreting data from the questionnaire. Such

categorical options included self employed and part time employed being stated as one response,

as well as married and engaged belonging to the same response option. The inability to

differentiate between these responses may have influenced the validity of the results in this

study. Additionally, 32 participants did not provide any response regarding current income level.

Therefore no premise can be drawn regarding the financial demographics of the sample.

Furthermore, as stress and burnout may be negatively denotative to individual perception, a

possibility remains that the titles of the questionnaires as well as questions regarding self-

reported stress levels may have evoked social desirability in participants, and led to an inaccurate

self-report of individual stress levels, as well as surreal reports of burnout prevention measures.

The burnout prevention assessment scores have not been displayed nor discussed in this paper, as

burnout prevention is not taken into consideration in the current hypothesis.

Conclusion

In conclusion, the results of the study displayed a higher score of stress and burnout in employed

participants in comparison to unemployed participants. However, as the mean values of the two

categories differed only by 1.0 (16.5-15.5), and as the effect size was calculated to be d=0.1312,

the generalizability of the results remains poor. Furthermore, the inability to distinguish between

self-employed participants and part time employed participants creates difficulty in data analysis,

as stress levels may differ from self-employment to part time employment. The increase of

student participants may also have led to misconceptions regarding actual stress levels of

employed and unemployed individuals. Therefore it is to be inferred that a poor correlation does

exist between employment status and increase of stress and burnout levels.

References

Anisman, H., Mantler, J., Matejicek, A. and Matheson, K. (2005) ‘Coping with employment

uncertainty: A comparison of employed and unemployed workers’. Journal of Occupational

Health Psychology [online] 10(3), 200-209. available from

<

http://web4.uwindsor.ca/users/f/fsirois/GradAppl.nsf/0/6c9e6979bbb25cac85256e1d006734ca/

$FILE/Mantler_etal_2005.pdf> [25 June 2015]

Bartley, M. (1994) ‘Unemployment and ill health: understanding the relationship’. Journal of

Epidemiology and Community Health [online] 48, 333-337. available from

<http://jech.bmj.com/content/48/4/333.full.pdf> [27 June 2015]

Castillo, L. G. and Misra, R. (2004) ‘Academic Stress Among College Students: Comparison of

American and International Students’. International Journal of Stress Management [online]

11(2), 132-148. available from

<http://www.johnbowne.org/ourpages/auto/2015/3/30/50025967/academic%20stress%20among

%20college%20students.pdf> [27 June 2015]

Cohen, S., Gordon, L. U. and Kessler, R. C. (1995) ‘Strategies for measuring stress in studies of

psychiatric and physical disorder’. cited in Cohen, S., Gordon, L. U. and Kessler, R. C. (n.d.)

Measuring stress. New York: Oxford University Press

Cohen, S., Janicki-Deverts, D. and Miller, G. E. (2007) ‘Psychological Stress and Disease’.

Journal of the American Medical Association [online] 298(14), 1685-1687. available from <

http://sites.northwestern.edu/foundationsofhealth/files/2013/03/07-JAMA-Psychological-stress-

disease.pdf> [27 June 2015]

Creed, P. A. and Macintyre, S. R. (2001) ‘The Relative Effects of Deprivation of the Latent and

Manifest Benefits of Employment on the Well-Being of Unemployed People’. Journal of

Occupational Health Psychology [online] 6(4), 324-331. available from <

ftp://190.25.233.21/DOCUMENTOS/Latitude/PDF/JOHP/2001(4)(324-331).pdf > [27 June

2015]

Deuschle, M., Haddad, L., Kirsch, P., Lederbogen, F., Meyer-Lindenberg, A., Pruessner, J. C.,

Rietschel, M., Schuch, P., Streit, F., Tost, H. and Wȕst, S. (2011) ‘City living and urban

upbringing affect neural social stress processing in humans’. Nature [online] 474, 498-501.

available from <http://www.nature.com/nature/journal/v474/n7352/abs/nature10190.html> [28

June 2015]

Fioroto, E., Losito, B. D., Miles, M. A., Simons, R. F., Ulrich, R. S. and Zelson, M. (1991)

‘Stress Recovery During Exposure to Natural and Urban Environments’. Journal of

Environmental Psychology [online] 11, 201-230. available from

<

http://www.researchgate.net/profile/Roger_Ulrich4/publication/222484914_Stress_recovery_dur

ing_exposure_to_natural_and_urban_environments/links/00b4953a3feb8bad67000000.pdf> [28

June 2015]

Kim, J. E. and Moen, P. (2002) ‘Retirement Transitions, Gender, and Psychological Well-Being:

A Life-Course, Ecological Model’. Journal of Gerontology [online] 57(3), 212-222. available

from <http://psychsocgerontology.oxfordjournals.org/content/57/3/P212.full.pdf> [27 June

2015]

Leiter, M. P., Maslach, C. and Schaufeli, W. B. (2001) ‘Job Burnout’. Annual Review of

Psychology [online] 52, 397-422. available from

<file:///C:/Users/User/Downloads/maslach_01_jobburnout.pdf> [27 June 2015]

Prottas, D. J. and Thompson, C. A. (2006) ‘Stress, Satisfaction, and the Work-Family Interface:

A Comparison of Self-Employed Business Owners, Independents, and Organizational

Employees’. Journal of Occupational Health Psychology [online] 11(4), 366-378. available

from <http://www.researchgate.net/profile/Cynthia_Thompson7/publication/

6737391_Stress_satisfaction_and_the_work-family_interface_a_comparison_of_self-

employed_business_owners_independents_and_organizational_employees/links/

53ebaa1c0cf250c8947a90a9.pdf> [25 June 2015]

Shields, M. (2006) ‘Stress and depression in the employed population’. Heath Reports [online]

17(4), 11-29. available from

<

http://www.researchgate.net/profile/Margot_Shields2/publication/6685427_Stress_and_depressi

on_in_the_employed_population/links/540b14ae0cf2d8daaabfbacc.pdf> [25 June 2015]

Tennant, C. (2001) ‘Work-related stress and depressive disorders’. Journal of Psychosomatic

Research [online] 51, 697-704. available from

<http://www.uic.edu/sph/glakes/occupational/reading_lists/73102/pdfs/tennant.pdf> [27 June

2015]