CHAPTER 3: DATA MANAGEMENT (Lesson 1: Data Collection)

Objectives
1. Determine the sample size from a given population
2. Perform sampling from target population
3. Use the different data gathering techniques
4. Present a project proposal to be studied using data management.

Data Management

Data come in different form and from different sources. These vast data are made available to assist us
in our decision – making. If these data are not properly managed and analyzed, everything will be meaningless
and void and lead us the false and unreliable information.

Statistics provide us the tool through which such data are collected, analyzed and presented to arrive at some
rich and interesting information. These tools which are derived from Mathematics, are useful in processing and
managing numerical data to describe the phenomenon and predict values.

Basic Concepts of Statistics

Statistics is a branch of science that deals with the collection, organization, presentation or tabulation,
analysis and interpretation of numerical data.
TYPES OF STATISTICS
 Descriptive Statistics refers to the processes that are used in presenting and describing data.
 Inferential Statistics refers to the processes of making inferences or predictions about a large group
(population) of data based on the observations of a smaller group of representatives (sample)

In the conduct of any study of data, one should not start collecting data without carefully defining a
population to be considered in the study. The population is the totality of the elements (person, objects, things
and animals) in which data is to be collected. If the population is large, collection of data can be costly and time
consuming. Hence, sampling is recommended. A sample is a representative of the population. It is also known
as the subset or subgroup of the population. The number that describes the population is called parameter
while the number that describes the sample is statistic.

Example, you might take a survey of all the million senior high school students all over the Philippines, but
because Philippines is a large population, so you take a sample. That may be a thousand senior high school
students in the selected region in the Philippines.

A substantial number of a sample is acceptable. Sample that is too large may lead to waste of money, time
and resources, while a sample that is too small may lead to inaccurate results. This inaccuracy is called as
sampling error.
The data from the sample is used to make inference about the population. So, how large should our
sample be? How closely we want our results to match the entire population. Finding a sample size can be one
of the challenging tasks. It depends upon the factors including the size of the population. Let us use a formula
to determine the sample size. If the population standard deviation is not known, use Slovin’s formula. To use
the formula, first figure out the margin of error “e”. A margin of error is an interval estimate. In a confidence
interval, it is the range of values below and above the sample statistic.

Determining Sample Size Using Slovin’s Formula

N
n= 2 , where N = population size; n = sample size; e = margin of error
1+ N e

Example 1: Find n, if N = 10,000 and e = 5%=0.05.

Solution:
N 10000 10000
n= = = =385
1+ N e 1+10000(0.05) 1+25
2 2

Activity 1
A. Solve the sample size, using Slovin’s formula
1. N = 20,000 and e = 5% n = _____________
2. N = 20,000 and e = 3% n = _____________
3. N = 20,000 and e = 1% n = _____________
4. What happened to the sample size if e is reduced?
5. Does the number of sample help in reducing errors in the study? Why?
Sampling Techniques
After computing for the sample size, our next concern is how to select the samples from the population. The
sample reflects the characteristics of the population from which it is drawn. Sampling is an act, process, or
technique of selecting appropriate sample, or representative of the population.

We use sampling rather than complete enumeration (census) because it is convenient and cheaper. Sampling
techniques are classified as either probability or nonprobability. Probability Sampling: Samples are randomly
chosen. Each member of the population has an equal chance to be selected. Nonprobability sampling:
Personal judgment plays a very important role in the selection. Each member of the population does not have
a known chance of being included in the sample.

Types of Probability Sampling

1. Simple random sampling: Each member of the population has an equal chance of being included in the
sample. It can use the table of random numbers and lottery or fish-bowl method.

Example, suppose the store owner wants to evaluate the performance of his staff. He writes all the names of
his staff in pieces of paper, and then draws 20 members who will be part of the survey.

2. Systematic Sampling: It selects every kth member of the population with starting point determine at
random. Where k = N/n. N – population, n - sample

Example, N = 100, n = 20, so k = 5. So, select every 5th member in the list starting at 5.

3. Stratified Random sampling: This is used if the population can be subdivided into strata. The samples can
be randomly selected from each stratum.

Example, a survey to find out if families living in a certain municipality with a population of 5000 are in favor of
excise tax. Suppose, you want to sample 370 families and ensure that all income groups are well represented,
respondents will be divided into income groups as shown below;

Income groups Number of Percent No. of sample

families
Upper class 1,000 20 % 0.2 (370) = 74
Middle class 1,500
Lower class 2,500
100% n = 370

This method of calculation of sample size from each stratum is called as proportional allocation.

4. Cluster sampling: This is usually used if the population is very large. It is sometimes called as area
sampling. The members of the population are divided into groups or clusters. (Geographical or Regional)

Example, we want to determine the average daily water consumption of families living in Malaybalay City.
There are 46 barangays in Malaybalay City. We can draw a random sample of 10 barangays using simple
random sampling and then a certain number families from each of the 10 barangays are chosen.

5. Multi-Stage Random Sampling – combination of two or more sampling methods are used

Activity 2. Below is the distribution of the students in a particular college. A training will be conducted outside
the campus, and the budget is good for 400 students only. Compute the number of students per program that
can go to the training.

Program Number of Percent No. of sample

students
Math 435
English 689
IT
Social Science 1,200
Total 2,881
Types of Nonprobability Sampling
1. Convenience sampling: Samples are chosen because of their availability. It also offers convenience to the
researcher.

Example, suppose you want to know the average income of the Filipino. You select the sample mobile phones.

2. Purposive sampling: It is selecting the sample on the basis of pre-determined criteria set by the
researcher. The sample is selected based on the objective or purpose of the study.

Example, the research is about the life of teenager who are already a mother. Of course, only teenage parents
will be the respondents.

3. Quota Sampling: This is the equivalent of stratified random sampling. The only difference is that the
selection of the samples in each stratum is not random.

Example, a survey is conducted to determine the most popular noon time show, each field researcher is
assigned to a certain area and given a quota of 300 viewers.

4. Snowball or Network Sampling: This is used if the desired sample is difficult to find or locate. This relies to
a referral, that is why, it is sometimes known as referral sampling.

Example, if one is interested with drug users, alcoholics or HIV positive. The researcher asks referrals from the
participants to locate potential samples.

Data Gathering Techniques

Characteristics measured from the person, object or thing is called as variable. The values of the
observations under a specific variable is called as data. Data can be classified as qualitative and quantitative.
Qualitative data are expressed in non-numeric such as categories, kinds, brands or names. Quantitative data
are expressed in numbers.

1. Direct method
 Observation – the researcher sees the situation directly to gather data. Researcher may use video
tape or audio tape.
 Interview – in-person or by telephone.
o Advantage: You can ask further questions for clarifications,
o Disadvantage: Person may not be available. Time consuming

2. Indirect method
 Questionnaire – a printed list of questions is used with or without the presence of the researcher. It
can be mailed or handed personally. It can done thru electronically aside from paper based.
o Advantages – you can gather data in a short period of time
o Disadvantage – some items will not be answered because they did not understand, costly or
expensive, Sometimes they will forget your questionnaire

Types of questionnaire
 Structured or closed format. The questions requires one answer only.
Ex. 1.) Are you in favor of k-12 curriculum? Ex. 2.) Which of the following brand of cellphone do
___ Yes __ No you like most?
___Samsung ___Nokia
___Cherry Mobile ___Oppo

 Unstructured or open format. The questions can have different answers. There are no limits as to the
responses of respondents.
Ex. 1.) What do you think are the reasons of bullying? 2.) Why did you choose this course?

 Rating Scale. It is often used to ask respondents to rate. One common example is the likert-scale.
Ex. 1.) The quality of education nowadays is improving. __SD __ D __N __A __SA

3. Registration method
This method gathers data from legal documents imposed by law. Data are sometimes obtained in
published or unpublished document. Age, sex and other information can be found in Philippine Statistics
Authority (PSA), number of registered cars can be obtained from Land Transportation Office (LTO), etc.