Collection Methods
Census and sample surveys essentially relate to the statistical collection of
data across various areas and sectors pertaining to the particular subject
matter or inquiry. These data collection exercises are undertaken on a
cross-section of a targeted population. The information that is derived
from the study of a population can be subsequently used for various
purposes.
The Census Method is also known as a Complete Enumeration Survey
Method. In this method each and every item in the universe is selected for
the data collection. The selected data might constitute a particular place, a
group of people, or any specific locality that is the complete set of items
and which are of interest in any particular situation. This method is most
commonly used by the government in connection with the national
population, housing census, agriculture census, etc. where vast knowledge
about these fields is required.
Census Method
The process used in the census method includes the statistical
compilation of all units or members of the target population under the
survey. In this case, population relates to the entire set of observations
connected to a particular study. For instance, if students of a university
have to give feedback on teaching faculty, the former will be held as the
population of that study.
Sample Method
Sample method chooses the different sample entities from the targeted
population. This method involves a statistical analysis of an already
determined number of observations that is derived from a larger set of
populations. Sample methodology can be used of different kinds; these
can be – simple random sampling or systematic sampling, cluster
sampling or stratified sampling, etc. among others.
Limitations of Census Method
● The expenditure incurred during the census is much higher because
of the sheer size of the population. Also, data is collected from each
unit of a sample population, which requires additional costing.
● Owing to the huge volume of data that is collated, a greater number
of the workforce (as well as man-hours) is required for completion.
● Costly Method: Census method is a very costly method of data
collection.
● Time Consuming: Census method consumes more time and labor
to complete data collecting tasks.
● Unsuitability: Census method is not applicable or suitable if the
universe is large. This method is suitable only for a small universe.
● Chance Of Errors: There is a comparatively higher chance of
statistical errors in this method.
Advantages of Sample Method
● In the sampling method, the number of units utilized is significantly
less. It helps to arrive at results much quicker.
● Sample method incurs substantially lower cost than the census
method as these tests are done over a limited sample.
● It has greater scope than the census method as it acts as a
substitute in such cases where the latter becomes impracticable. For
instance, if a manufacturer wants to test its range of toasters and
other kitchen appliances, it will apply a sample method and not a
census method.
● The nature of the sample method is such that it can be employed to
check the results from the census method. Also, due to the small
size of its sample, the method is useful for cross-checking the
reliability of its own results. A small sample can be taken out of
generated results, and that sample will have to be investigated.
● Accuracy And Reliability: Census method confirms a higher degree
of accuracy than other techniques. The Census method provides
complete information because each and every item is investigated
carefully. Therefore, it is a very reliable method of data collection.
● Suitability: Census method is effective if the universe is small.
● Intensive Study: Census method examines each unit completely
and gathers important data for intensive study.
● Indispensable: Census method is most reliable in certain cases
where other methods cannot provide reliable and accurate results.
● Heterogeneous Units: This method is also applicable to examine
heterogeneous units.
Sample Survey
A sample survey is a type of method that is used for collecting data from
or about the members of a population so that inferences about the entire
population can be obtained from a subset, or sample, of the population
members. A sample survey provides an estimate of the average length of
stays for surgical and nonsurgical discharges would be calculated and
compared. A well-defined sample survey will support inference from the
sample that is scientifically valid about the population.
Advantages of Sample Survey
● Low cost of sampling: The cost will be quite high if the data of
the sample survey is collected for the entire population. The
cost will be lower if data collected for sample surveys is for a
sample of population which is a big advantage.
● Less time-consuming in sampling: A sample survey takes less
time. A sample survey consumes less time than a census
technique.
● Scope of sampling is high: A sample survey makes it possible
to arrive at generalizations by studying the variables within a
relatively small proportion of the population.
● Accuracy of data is high: Drawing a sample and computing
the desired descriptive statistics, A sample survey makes it
possible to determine the stability of the obtained sample
value. A sample survey permits a high degree of accuracy due
to a limited area of operations.
● Organization of convenience: Organizational problems are
very few in sample surveys. The sample is of small size, vast
facilities are not required in it. The sample survey is therefore
economical in respect of resources. The study of samples in
sample surveys involves less space and equipment.
Limitations of Sample Survey
● Chances of bias: A sample survey involves biased selection and
thereby leads to drawing erroneous conclusions. Bias selection
arises when the method of selection of sample employed is faulty.
Comparatively small samples properly selected may be much more
reliable than large samples poorly selected.
● Difficulties in selecting a truly representative sample: Difficulties
in selecting a truly representative sample produce the most reliable
and accurate results only when they are representative of the whole
group. Selection of a sample is difficult when the phenomena under
study are of a complex nature. Selecting good samples for a survey
is difficult.
● Inadequate knowledge of the subject: The sampling method
requires adequate subject knowledge in sampling technique.
Sampling involves statistical analysis and calculation of error for
surveying. When the researcher lacks specialized knowledge in
sampling, they commit serious mistakes in sample surveying.
● The changeability of units: If the units of the population are not
homogeneous, the sampling technique will be unscientific. In this
sampling method, though the number of cases is small, surveying is
not always easy to stick to the selected cases. The units of the
sample may be dispersed widely.
Class Interval
Class interval is a term that is used to denote the numerical width of a class
in a frequency distribution. In a grouped frequency distribution, data is
arranged in the form of a class. The difference between the upper-class
limit and the lower limit gives the class interval.
In statistics, there are two types of class intervals, namely exclusive and
inclusive class intervals. Based on these, a frequency distribution table can
be constructed.
Definition of Class Interval
A class interval can be defined as the difference between the upper and
lower class limit of a frequency distribution table. In other words, it is the
width of each class. A grouped frequency distribution can be sorted either
on the basis of exclusive or inclusive class intervals.
Class Interval Formula
To determine the class interval, the lower limit of the class is subtracted
from the upper limit. The class interval formula is given as follows:
Class interval = Upper Limit - Lower Limit.
Inclusive Class Interval
An inclusive class interval is created using the inclusive method of sorting
data into a frequency distribution table. In such a method, the lower limit of a
class does not get repeated in the upper limit of the preceding class. When
a frequency distribution is analyzed the inclusive class interval has to be
converted to an exclusive class interval. This can be done by subtracting
0.5 from the lower class limit and adding 0.5 to the upper class limit. An
example of an inclusive class interval is given below:
Class Adjusted Class Frequency
10 - 19 9.5 - 19.5 2
20 - 29 19.5 - 29.5 5
30 - 39 29.5 - 39.5 1
Exclusive Class Interval
In an exclusive class interval, the upper limit of one class is the same as
the lower limit of the succeeding class. This method of sorting data ensures
continuity between two consecutive classes. The value of the mean, median,
and mode will be the same irrespective of whether the frequency distribution
table has an inclusive or exclusive class interval. Given below is an
example of an exclusive class interval.
Class Frequency
10 - 20 2
20 - 30 5
30 - 40 1
A Frequency Curve is a smooth curve which corresponds to the limiting case of a histogram
computed for a frequency distribution of a continuous distribution as the number of data points
becomes very large. Frequency curve is a limiting form of either histogram or frequency polygon.
There are four types of frequency curves which include bell shaped curve, U- shaped curve,
J-shaped curve and mixed curve.
A bell curve is a graph depicting the normal distribution, which has a shape reminiscent of a bell. ...
Its standard deviation depicts the bell curve's relative width around the mean. Bell curves (normal
distributions) are used commonly in statistics, including in analyzing economic and financial data.
The U-shaped curve usually refers to the nonlinear relationship between two variables, in
particular, a dependent and an independent variable. ... In medicine, U-shaped risk curves have been
found for risk factors such as cholesterol level, diastolic blood pressure, work stress, and alcohol
use.
A J shaped distribution is a probability distribution in the rough shape of the letter J placed on its
side, or its mirror image. These distributions tend to have some observations at one end, very few in
the middle, and a large number at the other end.
Simply put, a mixture distribution is a mixture of two or more probability distributions. Random
variables are drawn from more than one parent population to create a new distribution. The parent
populations can be univariate or multivariate, although the mixed distributions should have the
same dimensionality.