Absenteeism in work places is a critical problem affecting performance and competitiveness. In this study, absenteeism is defined as the absence of an employee at work, whether scheduled or not. In all workplaces, absenteeism falls into three categories: scheduled, unscheduled and partial shift absences. Scheduled absences consist of vacation or personal time, whether it is part of a paid time off (PTO) plan or separate. For such absences, excess costs arise when time is taken but not reported. Secondly, unscheduled absences include sick days, disability, family and medical leaves absence (FMLA), and workers’ compensation leave. Finally, partial shift absences include arriving at work late, leaving early, or taking longer breaks or lunches than allowed. The direct impact for FSL is reduced or delayed analytical work, or customers not being serviced. Unscheduled absenteeism is a chronic problem for FSL employers, conservatively costing very high unknown rates per salaried employees per year.

Absenteeism is a bigger problem in the FSL that managers have to face on daily routine [1] . The majority of employers have limited ability to accurately and regularly track how much absenteeism is reducing their bottom-line performance. This study uses the biometric data from a recently installed system to assess the extent of absenteeism among the FSL staff. Absenteeism can only be reduced if it can be measured accurately. The quantity used to estimate absenteeism is the absenteeism ratio; that is, hours absent over the total working hours expected expressed as a percentage. Reducing absenteeism will also help employers to better meet production and service demands without requiring an increase in headcount.

There is always little focus on absenteeism, compared to other costs like healthcare, for example, because healthcare is a unique line item in most organization’s budget, as compared to absenteeism, which rarely appears in any reports that corporate managers see on a regular basis. Moreover, healthcare costs are usually reported to management by external providers who maintain the data on more reliable systems than those used internally to track workforce-related costs. Equally, even amongst the largest companies, there is rarely a corporate manager who oversees the costs associated with absenteeism so that it goes unnoticed. Even in the FSL studied, the human resource officers are responsible for overseeing the data and not costs. Forensic science services are known for stressful working conditions and high rates of overtime with highest rates of absenteeism, necessitating quantification.

Many researchers have studied absenteeism in different areas with different approaches. An adequate approach to reducing absenteeism has been focused on medical absenteeism as this is the most prevalent form of absenteeism [2] . Others define absenteeism as a legitimate or illegitimate absence from school or class [3] . Absenteeism rates range differently amongst the workforce as long the employee is missing from work. In this study, absenteeism rates may be either high or low depending on the frequency of absenteeism in hours or days according to the period of study.

Workplace attendance and its converse absenteeism in any workplace are complex multi-dimensional issues involving the interaction and subtle interplay between employees, employer and its different departments leading to unknown impacts. In this study, employees in a workplace consisting of different laboratory disciplines, departments and units were studied. While the availability of resources such as reagents, facilities, instruments and equipment affect FSL performance, absenteeism plays a vital role, worth exploring [4] . The interactions between employees, employer and workplace conditions can have big effect on competitiveness [5] , as shown in Figure 1. Results of interaction between the three can affect the performance of the FSL via creation of case file backlogs, extended turnaround time (TAT), etc.

Figure 1 presents interplay between employees, employer and workplace with a critical focus on absenteeism, but can be extended to portray other aspects of interaction like engagement and job satisfaction [6] [7] [8] . Workplace attendance and absenteeism are of fundamental importance to FSL and the society it serves [9] .

Absenteeism adds to the total cost of operating the FSL and may reduce the quality, of FSL services. Consequences of laboratory-workplace non-attendance include but not limited to amplified costs, lesser morale, increased workloads and backlogging, non-achievement of objectives, reduced provision of services, decreased services quality, increased training costs and loss of key skills. Absenteeism is, therefore, a significant concern in the field of human resource management

which requires attention [10] [11] [12] . The prevalence of absenteeism in the FSLs necessitates research so that valuable contributions may be made that will, ultimately, empower management and positively contribute to the discipline being studied. The aim of this study was to measure or quantify absenteeism followed by investigating the contributing factors, and hence develop a framework for improving performance of the FSL [13] .

Whenever laboratory employees are absent, changes by the management have to be made to the usual operations. This may mean working the present employees longer and harder, allowing analytical processes to build up (i.e. backlogging) or be delayed (extended TAT), excluding certain services, hiring temporary replacements or permanently maintaining an extra workforce to cover absenteeism [12] .

The effects of absenteeism are most pronounced when the laboratory processes are interdependent, for instance, in processing criminal exhibits that require maintaining the chain of custody. FSL has complex, specialized and technological processes, operations of which will not proceed without competent staff and without legally authorized officers in all action points. The FSL has four departments and five units. All other units and departments (for instance, finance, human resource office, quality control coordination, procurement management unit, internal audit and ICT) work towards supporting the laboratory workflow. Absence of any employee in the laboratory and in support units impairs the performance of the laboratory as a whole.

As a result, laboratory and support cadre’s non-attendance leads to frustrated managers and supervisors, increased supervisory and administrative costs, adverse public perception and reduced confidence, of stakeholders [14] [15] . In addition, unseen costs such as additional supervisory time, extra staff, and overtime, and employment of temporary staff, result from absenteeism.

This study was possible after installation of a biometric time and attendance system within the FSL. Biometric time and attendance solutions have a direct impact on employee productivity increasing employee’s efficiency, accuracy and eliminating time stealing. Using a manual process for verifying the extra time the FSL has been working and reentry of data into the pay list, time required to process pay list for the entire FSLs were tedious and laborious in the past. The extent it takes to compile the time and attendance data through manual practices is significantly reduced to few minutes with a biometric time and attendance system. Also, productive hours are saved and the related personnel pay is reallocated elsewhere increases performance level through higher efficiency which in addition, when the biometric time and attendance registration and identification systems are integrated, they reduce and eliminate the need for any manual reentry of timekeeping data into a pay list system. This results in increased accuracy and reduced time in pay list processing and hence, another positive impact of biometric system.

Absenteeism data collection by biometric system is a process by which a person’s unique physical and other traits (such as facial marks, voice, finger scan, etc., are utilized to identify the employees. Such traits are sensed and recorded by an electronic device or system as a means of approving his or her identity. Biometric systems rely on specific data about unique biological traits such as the finger prints in order to work effectively. Of all the developing biometric technologies, fingerprint identification is one of the most distinguished and revealed biometrics for the reason of their distinctiveness and reliability over time [16] . Using biometrics have an advantage for FSL managers as it will only identify the specific employee by scanning his or her unique features such as fingerprint, iris or face. This will eliminate the problem of employees clocking in for one another, or needs to carry the card or remember passwords.

Absenteeism has a material effect on the bottom line of the organization with no exception of FSLs. In the absence of tracking tools and systems such as biometric systems, employers cannot adequately estimate their accrued liabilities, potentially creating an issue that impacts the performance of the FSLs. Before biometric system was installed, the attendance was marked manually in registers. Entry and exit times were unreliable and non-verifiable. Moreover, transferring register holding attendance records was an inefficient and highly time consuming process, making it difficult for laboratory managers to access it. However, employers such as in FSL have also introduced biometrics in the laboratories as a way of monitoring their staff [17] , making it easier to measure and quantify absenteeism.

In addition, laboratory managers have limited ability to accurately and regularly track to which extent absenteeism is reducing their bottom line competitiveness. This study provides a means of tracking absence for managers and supervisors. Reducing absenteeism will also assist the FSLs managers better to meet their expected turnaround time, reducing backlogging in sample analysis and case file management.

This study is part of the root cause analysis for productivity losses and competitiveness improvement framework, which is a critical step of determining corrective actions [18] [19] . In relation to quality management system, absenteeism is one of the non-conformity affecting the accreditation and certification process. Absenteeism may be considered as an unfaithful action, caused by either violation, routine or special causes. As far as violations are concerned, such as attending family issues, visiting the bank and late in to work or illegitimate breaks which are not reported, routine acts are those of extended tea and lunch breaks, whereby, employee need to have such breaks but are extended. Exceptional acts are those legitimate that are known, called for and are reported and always granted such as attending professional meetings, seminars, scene of crime visits and attending court sessions periodically. Other factors causing increased absenteeism are long work hours, increased workload and unsatisfactory work conditions [20] .

Many researchers have used different concepts to describe absenteeism. However, they all explain the absence of an employee to assigned work station no matter what reasons are given [21] . With the FSL, focus is on absence of employees at the headquarter laboratories premises, but absence from other work stations like inspection sites at ports of entry (PoE) and industrial premises, or sample collection areas are not scrutinized. Absenteeism has also been described as a concept of psychological problems or a form of withdrawal behavior when it is used as a means to escape an undesirable working environment [22] . Researchers further describe absenteeism as a challenge to corporate social responsibility, which impedes the efforts of the business to transform social problems into economic opportunities and economic benefits [23] . According to literature, absenteeism is also a measure of job dissatisfaction [6] [13] . As pressures increase on the budgets and competitiveness of the laboratories, more attention should be given to reducing workplace absenteeism and its related costs.

Organizations handle unscheduled absences in a number of ways. In some cases employees are simply not replaced, resulting in lost productivity. In other cases, organizations maintain excess staff, utilize overtime, or substitute with workers from specialized agencies. In any case, unscheduled absences are very disruptive to the business and result in increased indirect costs of absenteeism.

Case file processing in the FSL proceeds across several action points from receiving samples and requests to issuing approved reports, while maintaining the chain of custody. However, to complete case file processing other employees are required to support this system, absence of which affects the whole process. Lost productivity means that the organization runs too lean and production or service requirements are not met, leading to lost revenue paid by clients [14] [23] .

Overtime levels are higher in facilities with high absenteeism than in facilities with low absenteeism. FSL is one of the higher absenteeism organizations which require attention [24] . When excess staffing is practiced to overcome absenteeism, the headcount is higher than necessary in order to cover unplanned absences, all of which adds to the total running costs. This study focuses on the absenteeism of salaried employee as distinguished from hourly employees.

It is less usual for a salaried employee to be replaced when absent. Instead, the demands of customers (internal or external) are not met and depending on the employee’s position in the organization, the ability to create revenue or services delivery to clients is always affected. Compared to hourly workers, salaried workers are just as likely to be absent. Calculating the cost of absenteeism is more difficult for the salaried employees [25] . Excess absenteeism can also lead to increased healthcare costs, greater safety concerns and accidents (due to fatigue for workers in attendance), higher turnover and poor morale [26] . Therefore, total potential savings that could be achieved by reducing absenteeism and eliminating PTO abuse are significantly higher.

Safety issues arise in any workplace due to absenteeism, such that less-trained employees end up doing most of the work while tend to rush work after arriving late. Absenteeism is low in facilities with few fatigue issues (tired employees lead to more frequent and severe accidents). Poor health may result from absenteeism (if too much overtime is being used). Employees who work too many hours of overtime (which is associated with excess absenteeism) tend to get less sleep, and have a greater prevalence of cardiovascular and gastrointestinal disease, diabetes and musculoskeletal conditions [27] [28] [29] .

Understanding and defining workers absenteeism often receives only passing attention in more intervention-based studies. Absenteeism is an organizational problem that can be solved [9] [12] [15] . Moreover, there is surprisingly little up-to-date information on absence causes and controls available in a form that can be directly used by the FSL management.

However, the expected loss in daily productivity is incurred from employing a temporary substitute to replace a regular worker of average productivity. It is also significant that, daily productivity losses decline with the length of an absence spell due to increase in work load for the employees that continued working or taking over the part of work that the absent staff would have done. Different from this work, majority of organizations do not have automated means to track every instance of absence in one system and therefore lack the visibility necessary to address this critical business problem [12] [30] .

The data was collected through the process and the infrastructure shown in Figure 2. The infrastructure comprises of biometric figure detection system, camera (with IP address and staff photograph), network connections to server, PC at the data access point containing the Bio Star software (Version 1.62) to extract data from server. Figure 2 shows the biometric system setup used in this study for employee attendance data capture and analysis. The system comprised of fingerprint capturing device, digital image from finger print and the pattern-matching software converted into a code producing a digital image. Every time an employee punches the finger, the system performs quality check using pre-stored fingerprint features. On acceptance, the attendance details of the employee are stored in the server. Biometric systems rely on specific data about unique biological traits such as the finger prints in this case in order to work effectively. To start data analysis, a PC connected to the server via local area network (LAN) reads the database and transforms the data into MS Excel datasheet for analysis.

Table 1 shows the sample biometric data exported to the PC via options offered

by the Bio Star software, giving attendance details for nine selected staff identified by an identification numbers, for four months period. Based on user name information like gender, age, profession and skill levels were established. Biometric system was introduced by accommodating the attendee during the coming in and out when he winds up or closes. Therefore, the data analyzed in this study only calls for the individual working hours and days only. The total number of employees was 172 with some of employees in long study leave or maternity and sick leave. The data analyzed comprised of N = 134 staff which is about 78% of total number of staff. The data was collected for 12 months January to December 2016, which is a huge data assisted by use of automated system and server storage.

Overall absenteeism data collected was based on hours and days for all FSL employees. Absenteeism data was grouped according to departments, female and male gender [31] , time period of the year, professions (accountants, chemists and technologists), and also based on staff skill-levels (professionals, skilled, semi-skilled and unskilled levels). These are the factors assumed to affect absenteeism among a long list in literature [18] [19] [20] . For the purpose of this study, absenteeism was determined based on not being at the work station at that particular period of time, also regardless of reasons. Table 1 shows the details of the parameters assessed.

Since the biometric software yields total number of hours attended by a given employee, D_ath, then the hours an employee was absent, D_ah, were determined as per Equation (1):

D a h = D a t h − D w h (1)

where D_wh is the total number of hours an employee is supposed work for the specific period under study, that is, four months in this case. Similarly, given the total number of days attended by an employee in a given study period, D_atd, the number of days an employee was absent, D_ad, was determined as per Equation (2):

D a d = D a t d − D w d (2)

where D_wd is the total number of days an employee was supposed work for the specific period under study.

Based on the biometrics data for each individual employee, given the total number of hours an employee attended work, D_ah, and the total number of hours the given employee was supposed to work in the selected study period, D_wh, then absenteeism ratio based on hours, also expressed as a percentage [17] , was determined as per Equation (3):

A b h = D a h D w h × 100 % (3)

Similarly, absenteeism ratio based on days, A_bd, expressed as a percentage, was defined as a ratio between the numbers of days an employee is absent, D_ad, to the total number of days an employee was expected to be at work, D_wd in the period considered as per Equation (13):

A b d = D a d D w d × 100 % (4)

After establishing the absenteeism ratio, the cause factors of absenteeism were assessed and compared based on individual employees and also based on different factors affecting absenteeism as shown in Table 1. On the other hand, Table 2 summarizes the sample calculations for A_bh and A_bd, expressed as percentage of hours or days the respective worker was absent, for total of 656 hours and 82 days required for January to April, respectively, as per Table 1.

In this study, individual staff absenteeism is expressed as percentage absenteeism in days denoted as A_bd, or in hours A_bh. Figure 3 shows the overall percentage absenteeism in hours and day per individual staff identification number. It shows that A_bh is higher compared to absenteeism in days. This may be due to the employees absenting themselves during working hours, which is mostly coming late at work and leaving early. Therefore, the biometric system identifies or detects that the individual has arrived at work late and thereafter calculates

the total hours worked, stayed or present at work when the individual punches out.

Figure 3 shows also that the average value of A_bd is higher than A_bh. This indicates that employees often miss work for the whole day compared to missed hours of the working day. The values of absenteeism in days (A_bd) have the lowest value of 0% (there are no negative values, that is, either the individual came to work or not but not fractions of a day as easily depicted in hours). Negative values of A_bh means that this individual works more than the 8 hours of work day, that is, extra working hours which is frequently practiced in the FSL as a means of reducing backlogs. Thus, there are negative values for A_bh up to −50%, which is a good observation. The overall average value of A_bh = 13.69%, while that of A_bd = 20.46% for the months of May-August. Higher values of A_bh and A_bd may also be reflected as the causes of extended turnaround time (TAT) in laboratory processes which causes the FSL to have a reduced performance and competitiveness. The fact that A_bh < A_bd was observed for all three periods studied.

To study variations in the A_bh and A_bd, statistical analysis was conducted for different periods of January to April and May to August and September to December 2016, that is, at an interval of 4 months, regardless of gender, departments, skill levels, professions, etc., as shown in Table 3. The statistical analysis of overall A_bh data shows that the mean values of absenteeism ratio changed between the three periods of the year, that is, A_bh = 14.82%, 13.69% and 20.46%, for January-April, May-August and September-December, respectively, being highest in the September-December period. The lowest minimum values of A_bh were observed for January-April (A_bh = −66.91%) with the corresponding highest maximum values around 100% for all the three periods. On the other hand, the

A_bd data revealed higher but similar average values of A_bd = 21%, with higher minimum values of A_bd = −1.25%, 0% and 1.22% for the periods of January-April, May-August and September-December, respectively, compared to A_bh values. The maximum value of A_bd was highest for January-April compared to the rest of the periods of the year. In general, A_bd values are lower than A_bh values. That is, overall analysis shows that the average absenteeism ratio based on hours was lower than values based on days for the three periods studied. That is, employee’s absence is high on hourly basis than day basis. It is easier for employees to be absent for a part of the day (hourly absenteeism) than the whole day. Therefore, in terms of sensitivity analysis, hourly absenteeism is more sensitive than daily absenteeism. Negative minimum values of A_bh indicate that employees work for extra hours than for extra days, frequency of which has to be increased for improved competitiveness of the FSL.

Based on the values of standard deviation, there is an indication of strong variations in A_bh data than A_bd, due to higher values of the former. Since all data series indicates positive skewness, it is obvious that the absenteeism data is skewed towards the higher values of absenteeism ratio, also indicating a high tendency of absenteeism in the FSL, which require supervisors to investigate for root causes followed by improving or reducing A_bh. Higher kurtosis values for the A_bd data indicates that the PDFs of such data have high peaks than those of A_bh, showing high uniformity in the A_bd data, which concentrate around the mean values. There is a need for stringent control in attendance based on time, since proper management of time may help to reduce absenteeism by giving strict measures to a FSL staff that come in late frequently and leave early. A demarcation should be stipulated and agreed upon via an organizational staff attendance policy.

The decrease in overall average A_bh compared to A_bd can be attributed to a wide range of employees exhibiting very low values f absenteeism, A_bh below 0%, as shown in Figure 4. Also, for A_bh > 20%, the cumulative frequencies are highest for A_bd compared to A_bh, indicating large number of employees with high absence rates. On the other hand, Figure 4 shows the cumulative distribution

functions of the absenteeism ratios A_bh and A_bd. The two parameters differ in terms of span of values, where A_bh ranged between −50% and 100% while A_bd ranged between 0 and 71.7% only. Thus, A_bh is the suitable parameter for quantifying absenteeism due to its sensitivity, as the parameter allows assessment of slight changes in absenteeism.

In this study, the absenteeism rates were compared for male and female employees in the FSL during the period from January to April. The study involved 134 employees out of which 55 (41%) were female and 79 (51%) were male. Table 5 shows the statistics of the A_bh and A_bd data. The statistical data from Table 5 shows the average values to be A_bd = 20.36% and 21.10% for female and male employees respectively. The higher A_bh values for female employees can be attributed to breast feeding, pregnancy and maternity related absences. The purpose of the analysis was to determine if male and female employees had different patterns of absenteeism and how the two dynamics are affected by different factors. Although women perceived some work related factors differently than men and took substantially more hours off than men, their absence occurrences were not significantly different from men compared using A_bd. Even under presumably very similar employment and working conditions, women have 1.3 to 1.7 times as many absences as men [33] . Similar observations were reported in literature [11] [34] . In addition, Table 4 shows the values of standard deviation of A_bh and A_bd data (being higher for male than female employees) indicating that the absence rate for male employees vary more widely than female employees. Skewness of the absenteeism data shows similarity between female and male employees, while kurtosis shows that the PDF of A_bh is flatter than that of A_bd.

On average, female employees have high hourly absenteeism than male employees, while the observation was different for A_bd, whereby, male employees have high daily absence. That is, female employees work less hours per day, but more days than male employees in the FSL. Comparing the standard deviation values, it can be concluded that men’s absenteeism was less predictable compared to female employees, for both days and hours. Moreover, male employees were on the extremes of both hourly and daily absenteeism, that is, lowest minimum and highest maximum values with high skewness compared to female employees.

Figure 12 compares the A_bd data for male and female employees using PDFs. Both male and female employees have similar characteristics rate of absenteeism in days as shown in Figure 12 based on span of A_bd and shape of PDFs. Figure 12 shows also that the peak frequency of absenteeism rate, A_bd, for both male and female employees was observed at A_bd = 5%, whereas, the maximum values of A_bd ranged up to 100% for both male and female employees. Absenteeism data based on days may indicate travel days to attend duties outside the workplace like zonal laboratories and that male employees have external responsibilities outside the FSL more often than female employees. Such duties include crime

scene visits and court sessions for expert witnessing. This shows that gender difference strongly affects absenteeism rate in in the FSL.

Figure 13 presents a comparison between A_bh data for female and male employees using PDFs, showing the frequency distributions of A_bh data for female and male employees in the FSL are shown. With peaks at 5% and −5% for female and male employees, respectively, results show that female employees generally had higher rates of hourly absence. Results show that more female workers take time off work during work days more often than males. This observation is similar to literature reports which indicate that women have short spells of absence consistent with other studies comparing sickness absence in men and women [35] . A peak at A_bh = −5% for male employees indicates that they attend work for extra hours more often than female employees.

The span of A_bh data were the same for both female and male employees extending from −80% to 120%. However, as shown in Table 5, male employees are slightly on the extremes. Although A_bh values up to 120% were observed at low frequency, it is still important for the management to take prompt action especially for male employees. While the shapes of PDFs were similar for both male and female employees, the A_bh data shows higher peak 5% for female compared to male employees with peak at −5%. Also, with standard deviation of 34.24% and 39.02% for female and male employees, it indicates that the absence behavior of male employees is less uniform or unpredictable than for female employees, that is female employees have common absence behavior which is affected by common causes of absence while men’s absence are caused by large number of factors.

At this stage, absenteeism data was compared among skill levels to establish if

highly skilled or professional employees significantly exercise more absence than less skilled employees due to differences in nature of assignments. Figure 14 shows the analysis of average absenteeism data based on four employees’ professional levels, that is, professional, skilled, semi-skilled and unskilled. Observing more closely at how chronic absenteeism is defined and measured in these skill-levels, results indicate critical absenteeism among the professional cadre, being highest at A_bh = 25.62% on average during September-December. For the professional employees, A_bh was observed to increase during the three periods studied, but dropped considerably during the May-August period for semi-skilled and unskilled employees. This can be attributed to external assignments for the professionals and skilled employees compared to the unskilled employees. Financial year closes during May-August each year, requiring skilled and professional employees to travel to zonal laboratories for inspection, monitoring and evaluation purposes, and for attending budget presentation to the National Assembly. For all skill levels, absenteeism was highest during September-December, especially during end of the year which requires management action as more revenue is lost.

Therefore, it will be considered high absenteeism or low when the rate of missing at work in hours or days is related to value in terms of services offered or revenue collection. High absence rate for professional employees require scrutiny and possibility of directing the professional employees to produce high value output from whatever they do outside the FSL, as it impacts competitiveness. The high absence rate for professionals is again of high value only when individuals utilize their skills to bring benefit to the FSL when they attend external duties. Moreover, absence of semi-skilled and unskilled employees is also of value when they can acquire skills and experience useful to FSL during their absence. Because their operations are tied to the professional employees, semi-skilled and unskilled employees’ absence can disrupt productivity of the professionals,

for example, workflow relationship between a director and an office management secretary, or laboratory manager and analyst or laboratory attendant.

Figure 14 shows also that professionals had the highest absenteeism during all three periods studied, with the highest value of A_bh = 25.62% due to the fact that more external interaction is practiced by professionals than other skill levels such as skilled, semi-skilled and unskilled. Low absenteeism rate was observed in unskilled staff with a value of A_bh = -2.15% during May-August, attributable to the fact that such skill levels do not attend any external FSL issues. Most of the time, these unskilled staff perform their work by following stipulated daily instructions, being assigned instantly by their superiors. This skill level represents employees who clean, wind up analytical service areas, and put back in place all items such as cleaning tools, dustbins and washing instruments after analytical work in the laboratory. There are times when they spend extra time after working hours, and also during weekends and public holidays, leading to large number of hours worked and hence very low absenteeism ratio. The absenteeism data analysis based on skill levels, offers a unique nationwide relationship between the FSL and other organizations, a link created via professionals and skilled staff.

Figure 15 shows the PDFs of A_bh data for different skill-levels among FSL employees. The results show that skill-levels lead to dissimilar trends in absenteeism with respect to A_bh, that is, skill-levels influence the absence behavior strongly causing different shapes and span of PDFs. Only the skilled and semi-skilled employees showed lower values of A_bh beyond −60%. The peak location in the A_bh data varied between the skill-levels, with the unskilled staff

showing a peak at A_bh = −30%, while the rest of skill-levels show a peak at A_bh = 10%. It is interesting to note that there is extended absenteeism ratio up to A_bh > 100% observed for all skill levels, especially the semi-skilled employees, for which A_bh reached up to 120%.

Figure 16 presents analysis comparison of absenteeism data according to staff professions for the three different periods of the year. In this study, three different professions in FSL were selected, that is, accountants, chemists and technologists. Technologists and accountants are the main support cadres for the chemist profession. When A_bh data was compared between the three professions, the trend shows that absenteeism is most experienced among technologist which not only put burden on chemists, but also the quality of analytical services delivered to the clients is affected [1] .

While chemists and technologists work in the same place and conditions, hourly absenteeism varies among these professions (Figure 16) and fluctuates differently with time. Based on Figure 16, the technologists have the highest observed absenteeism rates of A_bh up to 43.48% (comparably the highest average value for all periods of the year), followed by chemists. The least values of A_bh were observed for accountants, being down to −35% for the periods of May-August. Results show that, the chemists and the technologists have high positive values of A_bh, indicating that they have more off laboratory work hours than the accountants who have negative values of A_bh. This shows that technologists arrive late at work and leave early.

Results from this study show that the accountants are present at work more than the expected 8 hours of work implying that they punch in early and exit

late, contributing to extended working hours. This is encouraging since payments preparations for internal and external clients are effectively done and on time, minimizing complaints. The lowest value of absenteeism observed among accountants for both period of the year, may also be caused by the type of work involved in the accounts section. Preparation of budget, balancing all the accounts and making sure that funds are available for either emergency or for the planned activities of FSL, are independent tasks, not requiring presence of other staff. As a results, such tasks can be done off-working hours for committed staff, leading to lower absenteeism ratio. On the other hand, the increase in absenteeism among technologists may be due to their job description, which is dependent on chemists. Moreover, the lower absenteeism ratio for chemists compared to technologists, can be attributed to the fact that chemists hold leadership positions, making them more responsible in terms of attendance and presence at work than technologists.

Table 5 shows the statistical analysis of the absenteeism data for the period of January-April, and also compares the average values of hourly absenteeism data for different professions within the FSL. The observed negative values of A_bh for the accountants (with mean of −4.94), compared to technologists and chemists (which shows mean values of A_bh = 36.58% and 12.58%, respectively), is also an indication of differences in the nature of work and required interactions among staff for completion of different tasks. However, higher A_bh values for technologists and chemists require action by the FSL management. While chemists are required to interact with staff from other sections of the FSL to accomplish their tasks, the case is different for accountants. Thus, absence among chemists affects strongly the performance of FSL.

Chemist is the profession that has the duty to prepare procedures and methods and supervise all the analytical work. If absenteeism is experienced among the chemist profession, it will be exhibited 3 times more among technologist, as shown in Table 5. This means, technologists have to wait for case files and instructions from chemists after which, data analysis and report writing and review is done by chemists. Where chemists have A_bh = 12.58%, the observed value

for technologist is 3 times, at 36.58%.

Figure 17 shows PDFs of hourly absenteeism data for different professions within the FSL. While accountants have the main peak at about A_bh = 15%, and a small peak around A_bh = −55% exhibiting bimodal behavior, technologists have a main peak shifted to the right at 30%, and a small peak further to the higher absenteeism values at A_bh = 90%. A peak at A_bh = −55% for accountants implies that this profession works extra hours more often than chemists and technologists. The main peaks for accountants and chemists are located at 10%. Both chemists and accountants have the tendency to work extra hours compared to technologists, as depicted in Figure 17.