Lesson 2 Methods of Data Collection
Lesson 2 Methods of Data Collection
Lesson 2 Methods of Data Collection
In any statistical investigation, there are a lot of ways of collecting data. Unfortunately, none of these
methods would become the best method because the choice of appropriate methods to be used largely
depends on some factors which include the definition of the problem, the research design, the time
element of data collection and the cooperation of the respondents.
1. The observation method is the simplest data collection technique. Here the data are obtained by
merely observing the behavior of persons, or objects but only at a particular time of occurrence. The
data obtained is called an observational data.
2. The experimental method is especially useful when one wants to collect data for cause- and- effect
studies under controlled conditions. In this method, there is actual human interference with conditions
or situations that can affect the variable under study. The data obtained in this method is called an
experimental data.
3. In the registration method, the respondents provide the necessary information in compliance with
existing laws. For instance, data can be derived from car registration, birth registration, voter’s
registration and marriage registration.
4. The use of existing studies also provides an archival method of data collection. In this method, the
primary source is the source in which the data are measured by the researcher or agency that publish it.
The secondary source or the other hand is the source from which any republication of data is made by
another agency.
5. In the survey method, the desired information is obtained through asking questions. The survey
method or either be direct or personal interview method and indirect or questionnaire method.
In the direct or personal interview method, there is a person to person contact between the
interviewer and the interviewee. This is considered as one of the most effective methods of data
collection because accurate and precise information can be directly obtained and verified from the
respondents. Moreover, this has a higher response rate but can be administered to the respondent’s
one at a time.
In indirect or questioner method is considered the easiest method of data collection, though the
use of questionnaire or a data gathered tool. Unlike the direct method, this method has a lower
response note but can be administered to a large number of respondents simultaneously.
Sampling Techniques:
Generally, data collection can be classified into two namely census and survey sampling. Census or
complete enumeration is a method of gathering data from every unit in the population. However, this is
not always possible because at times the population is unknown and therefore the population frame is
not always available. Besides, conducting a census entails a large amount of money, time and effort.
Survey sampling on the other hand is a method of gathering data from every selected sample in a given
population. In most cases, many researchers utilized survey sampling because it reduces cost; it has
greater scope, speed and accuracy.
Survey sampling can either be probability and non-probability sampling. Probability sampling is a
sampling procedure in which every element in the population has a known non-zero chance of being
included in the sample selection. The following are some common probability sampling designs.
a. Simple Random Sampling. This is a sampling procedure in which every element in the
population has an equal chance of being included in the sample. Here, we select n units out of N
units in the population, the selection of which is through lottery, raffle or use of random numbers.
b. Stratified Random Sampling. In stratified random sampling, the population of N units is divided
into sub-population called strata, which consists of more or less homogeneous units. Then from
each stratum a simple random sample procedure is performed in order to select the samples the
selection of which is independent in each strata. Thus, assure of homogeneity of units, a
stratification variable is first defined.
Example:
In a study to determine the student’s preference over socialized tuition fee at ABC University, a
researcher may employ stratified random sampling by stratifying all students of ABC University
(population) by college (stratification variable)
c. Systematic Sampling. This is a sampling procedure which selects the sample by taking every
N
kth unit from an ordered population. Here k is called the sampling interval and is defined as: k= n