3.1 Study area and period
The study was conducted in West Hararghe Zones on health workers in public hospitals [Chiro, Galemso and Asebot Hospitals]. West Hararghe zone is one of 20 zonal administrations in Oromia regional state located at 326km East of Addis Ababa on the way to Harar and Dire Dawa. Total populations of the zone are 2,582,209 with male 1,318,928 and females 1,263,281. The zone shares boundaries with East Shawa and Afar Regional state in North, East Hararghe in east, Arsi in west, Bale in south directions. West Hararghe zone is divided into 14 Woredas and 2 town administrations with Chiro as its capital. There are three public hospitals: Chiro, Galemso and Asebot hospitals.
Currently a number of health workers on job are Chiro hospitals have 275 health workers, Galemso hospitals have 256 health workers and Asebot have 139 health workers. Hospitals in the zones have total of 670 health workers according to sources from their human resources records. Health workers have two categories as health professionals 489 in numbers and supportive staffs 181 in numbers were eligible for this study. Both Chiro and Galemso hospitals are General hospitals and Asebot hospital is a primary hospital. The data study was conducted from August 27, 2018 to September 17, 2018.
3.2 Study design
A Facility based cross-sectional study design was used.
3.3.1 Source population
The source populations for the study were all health workers employed in public hospitals of West Hararghe zones.
3.3.2 Study population
The study populations were all randomly selected health workers in three hospitals [Chiro, Galemso and Asebot] in west Hararghe zones.
3.3.3 Inclusion and exclusion criteria
Health workers those permanently employed in three public hospitals and had greater than or equal 6 months‟ work experience were included. Health workers who were on the leave and long-term training during the data collection were excluded.
3.4 Sample Size and sampling technique
3.4.1 Sample Size determination
The sample size calculation is based on a single population proportion formula. N = [Z2a/2pq]/d2
Where n = number of sample size
p=proportion of health workers job performance, since there is lack of enough information of study done in this area on proportion of health workers toward their job performances we took 50%, 0.5=p
z= Z-score at 95% confidence interval = 1.96
d= Acceptable margin of error = 5% the formula for calculating the sample size [n] was: n= [Zα/2]2 p [1-P]
n= [1.96]2.0.5[1-0.5] Single population proportion formula
n = 384
„N‟ is the total number of health workers present in three public hospitals which means potential source population [N=670] during the study period „n‟ is the initial sample size without correction formula [if used the selected sample size were become lower and cannot get enough representative sample size] this could be taken as sample size by adding 10%of non-response rate [384*10%] +384 the final total sample size were reached 422.
3.4.2 Sampling technique
Simple random sampling method was used to select the study participants from each hospital. The total sample size required for the study were allocated to each of the three hospitals selected as proportional to the size method based on the number of health workers exist in each hospital [Figure 1].
3.5 Data Collection tools and Procedures
3.5.1 Data collection tools
Data were collected by using self-administered questionnaire which adopted from reviewing relevant literatures from websites, books and articles[27-30].
Job performance were measured by nine items using Likert scale with five responses [1=very poor, 2= poor, 3=good, 4= very good, and 5 = excellent]. Working environments consists eight items, working condition consists eight items and job satisfaction consists 11 items and all variable were measured by using Likert scale with five responses [1=strongly disagree, 2=disagree, 3=unknown, 4=agree, 5=strongly agree]. The questionnaires were prepared in English and translated to the local language Afaan Oromo and back to English by two independent persons from different place to check for consistency. These two
independent individuals had educational background of masters in English language and had experiences of teaching for long period of time.
3.5.2 Data collection method
The supervisor and data collectors were trained for three days intensively on the study objectives, the method of data collection, and the tools for data collection. Supervisor also took orientation on the way of successful data collection was achieved. Due to the nature of shift work in a hospital setting [off time duty] data were collected both during day and night. Completed questionnaires were checked every day by principal investigator
One health officer as supervisor and three nurses as data collectors were recruited from Tullo woreda Health office which is different from study area.
3.6 Variables for the study
3.6.1 Dependent Variables
3.6.2 Independent Variables
a) Socio-demographic factors
b) Working environments
c) Working condition
d) Job satisfaction
3.7 Operational definitions
i. Job performances defined as the effectiveness of the person in carrying out his or her roles and responsibilities related to direct patient Job performance is measurement of health workers performance in terms of efficiency, effectiveness, productivity and timeliness. Effectiveness is indicates the degree to which the process output [work product] conforms to requirements. Efficiency indicates the degree to which the process produces the required output at a minimum resource cost. The timeliness aspect measures on whether a unit of work was done correctly and on time. The productivity checks on the value added by the process divided by the value of the labor and capital consume. Job performance was
measured by nine items using Likert scale with five responses [1=very poor, 2= poor, 3=good, 4= very good, and 5 = excellent].The scores were averaged so as to show each participant‟s job performance level ranging from 9 to 45.
Poor job performance when respondent average score of job performance was less than computed overall mean value of job performance [Mean<3.56].
Good job performance when respondent average score of job performance was greater than or equal to computed overall mean value of job performance [Mean>=3.56].
ii. Working environment is a characteristics of the environment in which a person is expected to work. The Working environment includes physical, geographical location and social environment, learning opportunity, staff relation and benefits. It defined as environment that attracts individuals into the health professions, encourages them to remain in the health workforce and enables them to perform effectively with availability of essential materials and supplies.Working environment consist eight items, measured by using Likert scale with five responses. The scores were averaged so as to show each participant‟s working environment situation ranging from 8 to 40[1= strongly disagree, 2= disagree, 3= unknown, 4= agree, 5= strongly agree].
Bad environment when respondent with average score of working environment was less than computed overall mean value of working environment [Mean<3.07].
Good environment when respondent average score of working environment was greater than or equal to computed overall mean value working environment [Mean>=3.07].
iii. Working condition range from working time [hours of work, rest periods, and work schedules] to remuneration, as well as the physical conditions and mental demands that exist in the workplace. It also defined as commitment, workloads, working hours and others that affect job performances of health workers in health Working condition consists eight items which measured by using Likert scale with five responses [1=strongly disagree, 2=disagree, 3=unknown, 4=agree, 5=strongly agree]. The scores were averaged so as to show each participant‟s working environment situation ranging from 8 to 40.
Poor working condition when respondent with average score of working condition was less than computed overall mean value of working condition [Mean<3.36].
Good working condition when respondent average score of working condition was greater than or equal to computed overall mean value working condition [Mean>=3.36].
iv. Job satisfaction was measured by using the short form adopted from Minnesota Satisfaction Questionnaire[MSQ]. This instruments utilize 11 items and respondents rate their own job satisfaction level using a Likert scale with five responses [1=strongly disagree, 2=disagree, 3=unknown, 4=agree, 5=strongly agree].The scores were averaged so as to show each participant‟s satisfaction level ranging from 11 to 55. Respondents with average score of less than mean value were classified as dissatisfied, and those with average score of mean value and above were considered as
Dissatisfied when respondent with average score of job satisfaction was less than computed overall mean value of job satisfaction [Mean<3.19].
Satisfied when respondent average score of job satisfaction was greater than or equal to computed overall mean value job satisfaction [Mean>=3.19].
v. Health workers‟ refers to all employees of the
3.8 Data Processing and Analysis
Data were first checked manually for completeness, in consistencies and then entered through double entry into Epidata version 3.1 and finally exported to SPSS version 23 for analysis. Descriptive statistics such as frequency, percentages, mean and standard deviation were used to summarize the data and results were presented by using frequency tables and graphs.
Binary logistic regression was computed to see associated between each independent variables and level of job performance using crude odd ratio[COR] 95% C.I, variables which had p-value p<0.25 in bivariate analysis were taken to multivariate analysis in final models.
Multivariate logistic regression was computed to identify predictor variables associated with level of job performance using adjusted odds ratio [AOR] with 95% C.I, variables in the multivariable analysis with p
≤ 0.05 were considered statistically significant.
3.9 Data Quality Management
For ensuring the data quality before starting data collection, the questionnaires were pretested at Hirna primary Hospital by taking 5% of sample size. Then after, the necessary comments and feedbacks were incorporated, slight modification was done in the final tool. Training was given for supervisor and data collectors on the objectives of the study, contents of the questionnaires and process of data collection. Close supervision was carried out on daily basis by the principal investigator and the questionnaires were checked during the data collection process to ensure completeness and consistency of the information.
The reliability of questionnaires for each variable was measured using the Cronbach‟s coefficient alpha. The results of calculated Cronbach‟s coefficient alpha were obtained from job performance [0.836], working environment [0.855], working condition [0.737] and job satisfaction [0.816]. All variable had Cronbach‟s alpha value of 0.70 and above indicate that the items are reliable and considered good.
3.10 Ethical considerations
Ethical clearance was obtained from Jimma University Institutional Research and Ethical Committee and then official letter was obtained from Jimma University, Faculty of Public Health, Department of Health Economics, Management and Policy to West Hararghe Zone of Health Office. Zonal health office were wrote letter to all selected hospitals in zone. Data collectors were delivered to letter from zone and reach at employed hospitals and communicate with managers or responsible body of the hospitals. Finally data collectors were got permission and started contacting with each participant.
This study was conducted in accordance with the Declaration of Helsinki. Prior to administering the questionnaires, the aims and objectives of the study were explained to the participants and verbal consent was obtained from study participant after explaining the objective of study. They were also told that participation was voluntarily and confidentiality and anonymity were ensured throughout the execution of the study as participants were not required to disclose personal information on the questionnaire.
3.11 Dissemination plan
The findings will be presented to the Jimma University Institute of Health, Department of Health Economics, Management and Policy. The findings will be communicated to the human resource managers working on West Hararghe zones public hospitals.