Work Sampling

What is Work Sampling

Work samplingis the statistical technique for determining
the proportion of time spent by workers in various defined
categories of activity (e.g. setting up a machine, assembling
two parts, idleetc.).It is as important as all other statistical
techniques because it permits quick analysis, recognition,
and enhancement of job responsibilities, tasks, performance
competencies, and organizational work flows. Other names
used for it are 'activity sampling', 'occurrence sampling', and
'ratio delay study'
It is a process of making sufficient random observations of an
operators activities to determine the relative amount of time
the operator spends on the various activities associated with
the job.

The major goal of work sampling is to determine how long, or how

much of the work day, is spent on specific types of work.
Work sampling may identify the fact that certain operators spend a large
portion of their time waiting for work, or performing paperwork tasks, or
even performing activities that are not included in their job descriptions.
One of the basic foundations of statistical sampling theory is the concept
that the larger the sample size, the results will be more accurate.
In work sampling, a sufficient number of observations must be made to be
sure that the results accurately summarize the work performed. There are
statistical formulas to help determine how many observations should be
made.

Uses of Work Sampling

Activity and delay sampling (to measure the activities
and delays of workers or machines. Ex:percentage of
working
Perfomance sampling (to measure working time and
nonworking time of a person on a manual task, and to
establish a performance index or performance level for
the person during his or her working time.
Work measurement (under certain circumstances, to
measure a manual task, that is, to establish a time
standard for an operation.
It can also be used to check the standard time
calculated by other methods of work studies.

 Work sampling was initially developed for determining

time allocation among workers' tasks in manufacturing
environments. However, the technique has also been
applied more broadly to examine work in a number of
different environments, such as healthcareand
construction.More recently, in the academic fields of
organizational psychology andorganizational behavior,
the basic technique has been developed into a detailed
job analysismethod for examining a range of different
research questions

Characteristics of work sampling

study
The study of work sampling has some general characteristics related to the
work condition:
One of them is the sufficient time available to perform the study. A work
sampling study usually requires a substantial period of time to complete.
There must be enough time available (several weeks or more) to conduct
the study.
Another characteristic is multiple workers. Work sampling is commonly
used to study the activities of multiple workers rather than one worker.
The third characteristic is long cycle time. The job covered in the study has
relatively a long cycle time.
The last condition is the non-repetitive work cycles. The work is not highly
repetitive. The jobs consist of various tasks rather than a single repetitive
task. However, it must be possible to classify the work activities into a
distinct number of categories.

CONDUCTING A STUDY
It is recommended that a uniform procedure should be followed to perform a
work sampling study is to
1.

Establish the Purpose

First, the objective of the study should be established. Work sampling can
be used to determine an overall perspective on the work done.
2.

Identify the Subjects

Second, the people performing the task must be identified, i.e. general
office work is being studied with the objective of determining overall
productivity.

3.

Identify the Measure of Output

The third step in making the study is the identification of the measure
of the output produced or the types of activities performed on the jobs
being studied. This step is especially important if the objective of the
study is to measure productivity with the intent of setting a standard.
4.

Establish a Time Period

Fourth, the time period during which the study will be conducted must
be established. Starting and stopping points for the study must be
defined as well.
5.

Define the Activities

This step involves defining the activities that are performed by the
people under study. For example, the definition used in a machine
utilization study, including only the categories of working, idle, and
idle-mechanical breakdown.

6.

7.

Determine the Number of Observations Needed

After the work elements are defined, the number of observations
for the desired accuracy at the desired confidence level must be
determined. If a reasonable guess cannot be made, then a trial
study of perhaps 20 to 40 observations should be made to get an
estimate.
Schedule the Observations
Once the number of required observations has been determined,
either from appropriate statistical calculations or from tables, and
the actual observations must be scheduled. Typically, the analyst
will assign an equal number of observations each day during the
course of the study. For example, if 800 observations are required
and 20 work days are established as an appropriate observation
time, 40 observations should be recorded each day.
A random number table can be used to establish the random
times for each observation.

8.

9.

Inform the Personnel Involved

Before the study is actually performed, the personnel involved
should be informed about the objective of the study and the
methodology that will be employed.

Record the Raw Data

The next is the actual recording of the raw data. Although this
recording can be performed by anyone, it is desirable that a
trained analyst be employed.
It is also very important that the observations be made at
exactly the same location every time.

10. Summarize the Data

After the data have been collected, they must be summarized.

Simple example of work sampling

The work sampling procedure in its simplest form consists of
making observations at random intervals of one or more
operators or machines and noting whether they are working or
idle.
For example;
Working=36
Idle=4
Total=40 observations
Percentage of idle time=(4/40)*100=10%
Percentage of working time=(36/40)*100=90%
If one operator performed task for an 8 hour day. Operator was idle 48
minutes of the day (480*0,10=48) and was working 432 minutes of the
day (480*0,90=432).

The Theory of Work Sampling

The theory of work sampling is based on the
fundamental law of probability: at a given instant, an
event can be either present or absent.
Random sampling requires that there be no bias in the
sampling process. It is important that the concept of
randomness be understood and carefully followed in
work sampling studies.

The normal distribution

curve
The normal distribution curve is typical of
the kind of frequency distribution which is
of importance in work sampling because it
represents graphically the probabilty of the
occurence of certain chance phenomena.

Confidence level
At the outset it is necessary to decide what level of
confidence is desired in the final work sampling
results.
The most common interval is 95%
The area under the curve at 2 sigma or two standard
deviations is 95.45 %.
This means that the probability is that 95% of the time
the random observations will represent the facts and 5
percent of the time they will not.
The formula for determining the sample size for a
confidence level of 68 percent or 1 sigma is
S=desired relative accuracy, p=percentage expressed
as a decimal, N=number of random observations
(sample size)

Sp

p (1 p)
N

Accuracy of work sampling

measurement
For many kinds of measurement an accuracy of
+,-5 percent is considered satisfactory. This is
sometimes referred to as the standard error of the
percentage.
Following illustrations we will assume that a
confidence level of 95% and an accuracy of +,-5%
satisfactory. Also assuming that the binomial
distribution is used as the basis for determining
the error, the formula for determining the number
of observations required is
S=desired relative accuracy
p= percentage occurence of an activity or delay
being measured, expressed as a percentage of the
total number of observations
p (1 p) or as a decimal
Sp of2 observations (sample size)
N=Total number

Example
Suppose we want to determine the percentage of idle
time of the automatic screw machines in a department.
CL=95% error=+,-5 %
We want to know how many random observations will
be needed to give us the desired results.
100 observations were made and preliminary study 25
observations showed the machines to be idle. (25%)

Solution
N=? p=25% S=+,-5%=+,-0.05
p (1 p )
0.05 p 2
N

0.0025 p

4 p (1 p )
p (1 p )
4

N
N

4 p(1 p ) 4(1 p) 1600(1 p )

N

2
0.0025 p
p
0.0025 p

1600(1 0.25)
4800
0.25

Determination of Accuracy
Desired accuracy must be lower than 5%. For example,
if we calculates S is 3.5% then it is sufficient result.
The accuracy or standard error of +,-3.5 percent means
that the results is correct within +,-3.5 percent of ex:30
percent (+,-3.5%*30%=1.05%) or the value was
between 28.95% and 31.05 percent.
The 95% CL means that the probability is that in 95
cases of 100 the above results will represent the facts.

Determining Time Standards by Work Sampling

Sampling can be used for measuring work as well as for measuring delays, idle times
and performance.
Sampling can be used instead of time study for long activities.
It is possible to determine by work sampling the percentage of the day that a person
is idle and the percentage of the day that he is working, as well as the average
performance index or speed at which he worked the working portion of the day.
For example, 8-hour day as a drill-press operator. A work sampling study might show
15% idle or 72 minutes (480*0.15=72) and 408 minutes (480*0.85)
At an average performance index of 110 percent.
During the day 420 pieces

Determination of sample size

As well as defining the confidence level for our observations we have to decide on the margin of
error that we can allow for these observations.
Let us look at our example about the productive time and the idle time of the machines in a
factory. There are two methods of determining the sample size that would be appropriate for
this example:
the statistical method and the nomogram method.

Statistical method. The formula used in this method is:

 Let us assume that some 100 observations were carried out as a

preliminary study and at random, and that these showed the
machine to be idle in 25 per cent of the cases (p = 25) and to be
working 75 per cent of the time (q = 75).
We thus have approximate values for p and q; in order now to
determine the value of n.
Let us choose a confidence level of 95 per cent with a 10 per cent
margin of error (that is, we are confident that in 95 per cent of
the cases our estimates will be 10 per cent of the real value).

Nomogra
m
method
An easier
way to
determine
sample size
is to read off
the number
of
observation
s
needed
directly
from a
nomogram
such as the
one
reproduced
in figure 91.

Making random observations

To ensure that our observations are in fact made at random, we can use a random table such as
the one in table 12.
Various types of random table exist, and these can be used in different ways. In our case let us
assume that we shall carry out our observations during a day shift of eight hours, from 7 a.m. to 3
p.m. An eight-hour day has 480 minutes. These may be divided into 48 ten-minute periods.
We can start by choosing any number at random from our table, for example by closing our eyes
and placing a pencil point somewhere on the table. Let us assume that in this case we pick, by
simple chance, the number 11 which is in the second block, fourth column, fourth row (table 12).
We now choose any number between 1 and 10. Assume that we choose the number 2; we now go
down the column picking out every second reading and noting it down, as shown below (if we had
chosen the number 3, we should pick out every third figure, and so on).
11 38 45 87 68 20 11 26 49 05

 Looking at these numbers, we find that we have to discard 87, 68

and 49 because they are too high (since we have only 48 tenminute periods, any number above 48 has to be discarded).
Similarly, the second 11 will also have to be discarded since it is a
number that has already been picked out. We therefore have to
continue with our readings to replace the four numbers we have
discarded. Using the same method, that is choosing every second
number after the last one (05), we now have 14 15 47 22
These four numbers are within the desired range and have not
appeared before. Our final selection may now be arranged
numerically and the times of observation throughout the eight-hour
day worked out. Thus our smallest number (05) represents the fifth
ten-minute period after the work began at 7 a.m. Thus our first
observation will be at 7.50 a.m., and so on (table 13).

Example: Conducting the study

Determining the scope of the study. Before making our actual
observations, it is important that we decide on the objective of
our work sampling.
The simplest objective is that of determining whether a given
machine is idle or working. In such a case, our observations aim
at detecting one of two possibilities only:

We can, however, extend this simple model to try to find out the
cause of the stoppage of the machine:

Making the observations

So far we have taken the first five logical steps in conducting a
work sampling study.
selecting the job to be studied and determining the objectives
of the study;
making a preliminary observation to determine
approximate values of p (idle) and q (working);

the

in terms of a chosen confidence level and accuracy range,

determining n (the number of observations needed)
determining the frequency of observations, using random
tables;
designing record sheets to meet the objectives of the study.
There is one more step to take: that of making and recording
the observations and analyzing the results.

Work Sampling Observation

Form
Forms should be custom designed to accommodate the
specific data of interest to your study. Make the spaces
large enough to easily record the data. Provide
summary and calculations spaces right on the form.
Commercial software available for this application

Technique
Locate yourself at the same place each time prior to observing the
operation
Try to intentionally distract yourself from the variable of observation
as you approach the site (think safety)
Limit your time at the site to that actually needed for the
observation
Try to record only the minimum data that you will need to
successfully reconstruct the observation
Verify any discrepancy with the supervisor or foreman
Make notes on the form after the operator can no longer see you
Keep a pleasant attitude

Advantages of Work Sampling

Can be used to measure activities that are impractical
to measure by direct observation
Multiple subjects can be included
Requires less time and lower cost than continuous direct
observation
Training requirements less than DTS or PMTS
Less tiresome and monotonous on observer than
continuous observation
Being a subject in work sampling is less demanding
than being watched continuously for a long time

1. It does not require continuous observation by an

analyst over a long period of time
2. Clerical time is diminished,
3. The total work-hours expended by the analyst are
usually much fewer,
4. The operator is not subjected to long period
stopwatch observations.
5. Crew operations can be readily studied by a single
analyst

Disadvantages and Limitations

Not as accurate for setting time standards as other work
measurement techniques
Usually not practical to study a single subject
Work sampling provides less detailed information about
work elements than DTS or PMTS
Since work sampling deals with multiple subjects,
individual differences will be missed
Workers may be suspicious because they do not
understand the statistical basis of work sampling

Some Advantages Work Sampling in Comparison with Time

Study
Many operations or activities which are impractical or costly to
measure by time study can readily be measured by work
sampling.
A simultaneous work sampling study of several operators or
machines may be made by asingle observer. Ordinarily an
analyst is needed for each operator or machine when continous
time studies are made.
It usually requires fewer man-hours and costs less to make a
work sampling study than it does to make a continous time study.