Chapter IV: Data Management - Collection of Data

Data Gathering
Types of Data
1. Primary data are information collected from an original source of data, which is first-hand in nature. Primary sources
include government offices, private organizations, business establishments and recognized individuals who have first hand
information about an event.
2. Secondary data are information obtained from published or unpublished sources like newspapers, magazines, journals,
books and theses and other republished materials

Methods of Collecting Data

1. Direct or Interview Method. This method is considered as one of the most effective methods of collecting data. In this
method, there are at least two persons (an interviewer and interviewee/s) exchanging information. The researcher makes direct
and personal contact with the interviewee. The interviewer gathers data by asking the interviewee series of questions.
Advantages:
1. It can give complete information needed in the survey/study.
2. It can give precise and consistent information because clarification can be made. (The interviewer can immediately
clarify any misinterpretation made by the respondents. Also, the interviewer could repeat the question/s not fully
understood by the respondent.)
Limitations:
1. It is more expensive and time consuming.
2. It may yield inaccurate information. (The interviewer can influence the respondent’s answers through his facial
expression, tone of voice, or wording of the questions.)
3. The interviewer may turn in dishonest responses if his expected or desired responses are not obtained.
2. Indirect or Questionnaire Method. This method is one of the easiest methods of data gathering. It takes time to prepare
a questionnaire because it needs to be attractive. It can include illustrations, pictures, and sketches. Its directions must be precise,
clear and self-explanatory.
Advantages:
1. It is less expensive since questionnaire can be mailed or hand-carried.
2. It is less time-consuming since it can be distributed over a wider geographical area in a shorter time.
3. It can give confidential responses since the respondents can answer the questionnaire privately.
4. The answers are free from any influence from the interviewer.
Limitations:
1. It cannot be accomplished by illiterates.
2. It has a high proportion on non-response or non-retrieval. (Some respondents might ignore answering the
questionnaire.)
3. It tends to give incomplete or wrong information.
Two forms of Interviews and Questionnaires
1. Standardized or Structured. All questions are presented with exactly the same wording in the same order to
all subjects.
2. Unstructured. In this form, neither the questions nor the responses are determined before the interview.
Two forms of Questions
1. Fixed-Alternative Question allows only a fixed response, encourages a short or single-word answer.
Example: How often do you go to the library?
O once a day O when there is assignment
O every day O every other day
O once a week O never at all
2. Open-Ended Question allows a free response, encourages a meaningful answer from the respondent’s own
knowledge and/or feelings.
Examples: 1) What do you think can be done to reduce crime?
2) What are the common problems you encounter in studying Statistics?

Other Methods of Collecting Data

1. Observation Method is used to gather data regarding attitudes, behavior, values and cultural patterns of the subjects
under investigation. The researcher may observe subjects individually or group of individuals to obtain data and information related
to the objectives of the investigation. He may even relate or participate with his subjects in activities while doing the observation.

2. Experimentation Method. This method of collecting data is used to find out the cause and effect relationship of certain
phenomena or event under controlled conditions. An example is when a chemist is interested in finding out the effect of pesticide
when sprayed in vegetables which are then eaten by people.

3. Registration Method. This method of collecting data is governed by our existing laws. The researcher gathers data from
offices concerned, e.g. the Philippines Statistic Authority(PSA) (formerly the National Statistics Office (NSO)), the Commission of
Elections (COMELEC), Municipal/City Halls or Barangay Offices. The NSO keeps complete records of birth and death of the population
while the COMELEC takes the list of registered voters.

19
 4. Texting Method. The researcher may ask or invite individuals to send text opinions on certain issues or send in their
choices on their brand preferences on a particular product using their cellphone.

Determining the Sample Size

Most surveys conducted are done on a sample basis because of time and cost involved if the population is used. We use
the Slovin’s formula to determine the statistically acceptable sample size to be extracted from the given population. The Slovin’s
formula is

N
n = where n = sample size; N = population size ; e = margin of error
1+ N e2

The margin of error is the error we expect to commit in getting the sample. The sample is only the estimate of the
parameter.

Example 1: A group of student researchers is conducting a survey to find out the opinion of the residents of a particular Barangay
regarding the oil price hike. If there are 8000 residents in that Barangay and the student researchers decided to use a
sample with a 10% margin of error, what should be the sample size?

Solution:

N 8000
Given: N = 8000 e = 10% or (0.10) or (0.1) n= 2
=
1+ N e 1+ ( 8000 ) ¿ ¿

Example 2: Suppose in Example 1, the student researchers would like to use only a 5% margin of error, what should be the sample
size?

N 8000
Given: N = 8000 e = 5% or (0.05) n= = =380.95 = 381
1+ N e 1+(8000)(0.05)2
2

It is very clear from example 1 and 2 that as the margin of error decreases the sample size increases.

In some instances, researchers wanted to delimit their sample size due to budget and time constraint. With their desired
sample size, researchers would like to estimate their margin of error. Algebraically manipulating the Slovin’s formula, the margin of
error can be estimated as follows:

N −n
e=
√ nN
where n = sample size; N = population size

Example 3. From a population of 800, a researcher of limited resources decided to use only a sample of 50 respondents, estimate
his margin of error.

Solution:
N −n 800−50
e=
√ nN
=
√
50(800)
=0.1369∨0.14 or 14%
This would mean that if the researcher can afford only a sample of 50 respondents, she would have a
margin of error of about 14%.

Sampling Techniques

Sampling is a process of choosing a representative of a population.

Sample is the chosen representative of a population.
A population is the totality of objects, individuals, or things under consideration/study.

Two kinds of selecting/drawing samples

1) Probability Sampling(or random sampling) is a sampling procedure where each member of the population must have an
equal chance of being included in the sample.

2) Non-Probability Sampling is a sampling procedure where not all of the members of the population are given an equal
chance of being chosen for the sample. With these method, some members of the population may be deliberately ignored (that is,
giving them no chance to be included) in the choice of members for the sample.

Five kinds of Probability Sampling

1. Simple Random Sampling. This is the simplest form of random sampling. It is also known as lottery or fishbowl
technique. To illustrate this method, consider a population of 50 students. Each member of the population is numbered on a piece
of paper. The pieces of paper shall be identical (equal in size and weight) and rolled evenly. They are placed in a lottery box and
shaken very well. With eyes away from the rolled papers, the desired number of samples, 10 sheets (representing the 10

20
representative students) are drawn. If the population is large, lottery is cumbersome. A simple and practical way is using the
printed Table of Random Numbers found in one of the appendices of Statistics book or using the Random Numbers generated
through computer or scientific calculators.

2. Systematic Random Sampling is a process of selecting every nth or kth element in the population until the desired
sample size is obtained.

Example: Given N = 250 students; n = 30

N 250
Step 1: determine k (sampling interval) = = =8.33∨8
n 30
If k = 8, then every 8th element of the given population is selected until the desired sample is obtained.

Step 2: determine the random start by selecting one number from the numbers 1, 2, 3, 4, 5, 6, 7 and 8
( k = 8) by lottery sampling. If 4 happens to come out, then the random start is 4.

random start: 4 (the 4th member of the population is included in the sample)
2nd number is: 4 + k = 4 + 8 = 12 (the 12th member of the population is included in the sample)
3rd number is : 12 + k = 12 + 8 = 20 (the 20th member of the population is included in the sample)
4th number is: 20 + k = 20 + 8 = 28 (the 28th member of the population is included in the sample)
repeat the procedure until the desired sample size is obtained

Here are the 30 samples: 4, 12, 20, 28, 36, 44, 52, 60, 68, 76,
84, 92, 100, 108, 116, 124, 132, 140, 148, 156,
164, 172, 180, 188, 196, 204, 212, 220, 228, 236

3. Stratified Random Sampling is a process done by dividing the population into strata or categories and choosing the
members at random proportionate to each stratum or sub-group. This technique is particularly useful in populations that can be
stratified into groups by gender, race, or geography. This is done to avoid the error of selecting too few or too many members
from each stratum. The results are then grouped together to form the sample.

Example: Given: N = 3500 students; Male Students = 1500; Female Students = 2000
Use stratified random sampling to draw 100 students.

n 100
Step 1. determine p = = =0.02857∨0.0286
N 3500

Step 2. 1500 x 0.0286 = 42.9 = 43; 2000 x 0.0286 = 57.2 = 57

Group N p Sample Size

Male 1500 0.0286 43
Female 2000 0.0286 57
Total 3500 100

4. Cluster Sampling is a suitable procedure if the population is spread out over a wide geographical area. It is also a
practical sampling technique to use if the complete list of the members of the population is not available. This is sometimes called
area sampling.

A cluster is an intact group possessing a common characteristic.

Example: Given: N = 300 Third Year CIT students (3-A = 48, 3-B = 54, 3-C = 50, 3-D = 47, 3-E = 49, 3-F = 52)
number of sections = 6
300
average number of student per section = 50 (from =50 ¿
6
desired sample size = 100

100
Solution: Required Clusters = desired sample ¿ ¿ average number of student per section= =2 sections ¿
50

The two sections will be selected from six sections by lottery sampling. Suppose, the selected sections are 3-B and
3-E, then consider all students in these sections as members of the sample. Probably it would be more or less than 100
students.

21
 5. Multi-Stage Sampling is a more complex sampling technique. We use combinations of several random sampling
techniques in drawing the sample from a very large population. This is done by dividing the whole population by area, and then
each area into strata. Thereafter from each stratum, we get the sample by using the simple random sampling technique. For
example, a survey is to be conducted on the performance of Public Senior High School Teachers in the province of Bulacan.
Randomly select/draw the municipalities in the province. Randomly select the schools in the selected municipalities and then
randomly select the sample(senior high school teachers) from the selected schools.

Three Kinds of Non-Probability Sampling

1. Quota Sampling. A sampling technique used popularly in the field of opinion research. Selections are done using quota.
For example, a researcher decided to interview 30 avid fans during the PBA championship game at the Philippine Arena. At the start
of the game, he interviews the avid fans until he reached his quota of 30 and continues watching his favorite team compete the
game.

2. Purposive Sampling is done when the subject satisfies the criteria lay down by the researcher. It is also called as
subjective or selective sampling. In some cases, the researcher gets his sample from the respondents purposely related or close to
him. For example, a researcher wanted to find out how the drug addicts did able to overcome the treatment they experienced
inside the rehabilitation centers. The researcher could possibly choose to interview only those who have completely recovered or
managed to change their life to become better after rehabilitation.

3. Incidental/Accidental/Convenience Sampling. This is very popular in market research. In this sampling technique,
samples may be individuals in a restaurant, in a supermarket, in a department store, or in a shop, who are simply or readily available
in a convenient way to the researcher. For example, a researcher wanted to monitor the acceptance of a brand of laundry soap to
the buying customers. The researcher interviewed every customer who bought the product from the supermarket within a random
period of the day. Those who were interviewed are what we called incidental/accidental/convenience samples.

22
 Score
MMW101: Activity 8
2nd Semester, AY 2019-2020

Name: ______ Course/ Yr. / Sec. /Group: __________ Date: ____

Surname First Name M.I

I. Identificaton. Write the word or group of words on the lines provided before the number.

_________________________ 1. It is sometimes called as area sampling.

_________________________ 2. A process of selecting a representative of a population.
_________________________ 3. An information collected from an original source of data.
_________________________ 4. It is an intact group possessing a common characteristics.
_________________________ 5. A sampling procedure where not all of the members of the population are given an
equal chance of being chosen for the sample.

II. Identify whether the source of data is Primary or Secondary.

_________________________ 1. Annual sales and profits of a certain company downloaded from the company’s
website.
_________________________ 2. The Barangay Risk Reduction Funds read in the newspaper.
_________________________ 3. The files of the COMELEC on registered voters of Barangay Bulihan.
_________________________ 4. The news on Prayer Brigade for healing and protection as reported by a field reporter.
_________________________ 5. An article from a magazine about ‘Best Breakfast Smoothies for Diabetics.’

III. Identify the type of sampling used in the following statements.

_________________________ 1. A researcher interviews every 7th patient in the list of Bulacan Provincial Hospital in-
patients.
_________________________ 2. A student-researcher at BulSU surveys all students from each of 5 randomly selected
classes.
_________________________ 3. A company manager interviews supervisors on how can employees wellness program
be improved. He determines the proportionate allocation of 30 sample from fifteen
departments.
_________________________ 4. A class mayor received 3 Eat all You Can free meal tickets to be given away. She
wrote the name of each of her 40 classmates on a paper and placed them in a paper
cup. She picked 3 rolled tickets from the paper cup.
_________________________ 5. A certain company who produces a brand of coffee wanted to monitor the acceptance
of their product in the market by making phone calls using the phone numbers found
in the telephone directory.

IV. Determine the sample size. Show your solutions.

1) N = 15,000; e = 5% 2) N = 4,000; e = 12%

23
V. From a population of 3,500, estimate the margin of error for each sample size. Show your solutions.

1) 200 2) 350

VI. Determine the sample for each of the following problems.

1. AJT interior design management is interested to know about their 785 clients satisfaction to improve their
service. The firm decided to interview 30 clients. List the number of 30 samples. For the sake of uniformity,
let the random start be 6.

N
Step 1. k =
n

Step 2.
1) 6 6) 11) 16) 21) 26)
2) 7) 12) 17) 22) 27)
3) 8) 13) 18) 23) 28)
4) 9) 14) 19) 24) 29)
5) 10) 15) 20) 25) 30)

2) The College of Engineering conducts a survey on its graduating seniors in order to know the contribution of
the program to their personal and professional growth. Determine the sample for each major using
proportionate allocation. Use the given data in Table 1.

Table 1
Major N P n
Civil Engineering 85
Electrical Engineering 93
Computer Engineering 115
Mechanical Engineering 78
Electronics and Communication Eng’ng. 120
Total 490 /////////////////// 80

n
p=
N

24
25