Mod 2- Res Stat

(Found Ed 203)

Dr. Agapito P. Pino, Jr.

Professor
Module 2. On Sampling and Weighted Means

Name _______________ Date _______

Address _________________________
Contact number ______________
Content:
1. Sampling and Sampling Techniques
2. Sample size and the Slovin's Formula
3. Stratified Sampling
4. The Weighted Means
The Slovin's Formula

Sample Size
To determine a sample size from a given population needs an equation called
the Slovin's Formula. The equation is as follows:
n = N ÷ (1 + Ne2)
Example: Suppose you have a population of 4,000 cases and you feel that
it is too heavy for you to handle all of them not to mention the expenses
involved in doing the survey. You have to take only a portion of the
population. Observe that
n = 4,000 ÷ (1 + 4,000 x (0.05)2)
n = 4,000 ÷ (1 + 4,000 x 0025)
n = 4,000 ÷ (1 + 10)
n = 4,000 ÷ 11
n = 363.6 or 364 , the minimum number of cases that will make
your sampling reliable.
This means we will have to take at least 364 or more cases from the
population of 4,000 cases. Now, let us take a look at the Sampling
Techniques.
Sampling: Probability and Non-probability

In Statistics, survey sampling describes the process

of selecting a sample of elements from a target
population to conduct a survey. The term survey may
refer to different types or techniques of observation. A
sample survey therefore is a survey which is carried out
using a sampling method i.e. in which a portion only and
not the whole population is surveyed.
Now, how do you pick the 364 respondents from
4,000. This brings us to the different sampling
techniques. They are classified under two categories
namely probability sampling and non-probability
sampling. We will see which among those techniques is
most appropriate. Let us take a look at probability
sampling.
 Probability Sampling

Probability Sampling refers to those

sampling where all the elements in a
population have equal chances of being
selected.Sampleshere are representatives
of the elements in the population.
The following are the various methods
of probability sampling: simple random
sampling, systematic sampling, stratified
sampling,cluster sampling, stage sampling
Simple random sampling

Here each member of the population under

study has an equal chance of being selected.
The method involves selecting at random from a
list of the population (a sampling frame), the
required number of subjects for the sample.
Because of probability and chance, the sample
should contain subjects with characteristics
similar to the population as a whole; some old,
some young, some tall, some short,some fit,
some unfit, some rich, some poor,etc. One
problem associated with this particular sampling
method is that a complete list of the population
is needed and this is not always readily
available.
Systematic Sampling

This method is a modified form of simple

random sampling. It involves selecting subjects
from a population list in a systematic rather than
a random fashion. For example,if from a
population of say, N=2,000, a sample n=100 is
required, then every kth person (k= N/n) that is
k=2,000/100 = 20, can be selected as a sample.
The starting point for selection is chosen at
random. So then every 20th subject is chosen
as a sample.
Stratified Sampling

Stratified Sampling involves dividing the

population into homogeneous groups, each
group containing subjects with similar
characteristics. For example, group A might
contain males and group B females. In order to
obtain a sample representative of the whole
population in terms of sex, a random selection of
subjects from group A and group B must be
taken. If needed, the exact proportion of males
to females in the whole population can be
reflected in the sample.
Cluster Sampling

When the population is large and widely

dispersed, gathering a simple random sample
poses administrative problems. Suppose we
want to survey children's fitness levels in a
particularly large community. It would be quite
impractical to randomly select children and
spend an inordinate amount of time travelling
about in order to test them. By cluster sampling,
we can randomly select a specific number of
schools and test all the children in those
selected schools.
Stage Sampling

Stage sampling is an extension of cluster

sampling. It involves selecting the sample in
stages, that is, taking the samples from the
samples. One type of stage sampling might be
to select a number of schools at random, and
from within each of these schools, select a
number of classes at random and from within
these classes select a number of students.This
sampling technique is rarely used because of
the complexity of its strategy. It incurs a lot of
effort, time, and expense.
Non-probability Sampling Techniques

Small-scale surveys often resort to the use of

non-probability samples because despite the
disadvantages that arise from their non-
representativeness, they are far less
complicated to set up, they are considerably less
expensive, and can prove perfectly adequate
where researchers do not intend to generalize
their findings beyond the sample in question or
where they are simply piloting a survey
questionnaire as a prelude to their main study.
The chief kind of non-probability sampling are as
follows:
Convenience Sampling

Convenience Sampling or as it is
sometimes called accidental sampling -
involves choosing the nearest individuals
to serve as respondents and continuing
that process until the required sample size
is obtained. Captive audiences students,
teachers, or employees often serve as
respondents in surveys based upon
convenience sampling.
Quota Sampling

Quota sampling has been described as the

non-probability equivalent of stratified sampling.
It attempts to obtain representatives of the
various elements of the total population in the
proportions in which they occur there. Thus,
researchers interested in race relations in a
particular community might set a quota for each
ethnic group that is, proportionate to its
representation in the total population in the area
under survey.
Purposive Sampling

In purposive sampling, researchers

handpick the cases to be included in the
sample, on the basis of their judgement of
typicality. In this way, they build up a
sample that is satisfactory to their specific
needs
Dimensional Sampling

Dimensional sampling is simply a further refinement of

quota sampling. It involves identifying various factors of
interest in a population and obtaining at least one
respondent of every combination of those factors. Thus
in the study of race relations for instance, cited under
quota sampling, it must be noted that within each ethnic
group, we must distinguish between the attitudes of
recent immigrants, those who have been in the country
for some period of time, and those members of the
ethnic group who were born in another country. Their
sampling plan might take the form of a multi-dimensional
table with "ethnic group" across the top and "length of
stay" down the side.
Snowball Sampling

Researchers identify a small number of

individuals who have the characteristics
that they require. Those people are then
used as informants to identify others who
qualify for inclusion and these in turn, yet
identify others hence the term snowball
sampling.
 Let us now see how data are organized
when the research tool used is a
questionnaire or an opinionaire where
research respondents are asked to check
certain options for every item. Options like
Strongly Agree, Agree, Undecided,
Disagree, Strongly Disagree maybe
checked. Another set of options maybe
Always, Often, Sometimes, Seldom,
Never.
 In other words when you will undertake a
descriptive research and you use a
questionnaire or an opinionaire to gather
data, you just don't simply stop at tallying
frequencies, or ranking scores. You have
to compute weighted means. The reason
why you call them weighted means is
because you give weights to ranks or
options.
The Weighted Mean

Suppose a certain item in your questionnaire asks

the respondents to decide whether the implementation of
a school policy is effective in imposing discipline on
students. You were asking them to choose from among
the following options: Very Effective (VE), Effective (E),
Less Effective (LE), Not Effective (NE), or Don't Know
(DK). Then, you tally all the responses of the, say, 200
respondents who answered that specific item on the
questionnaire.
What could be the weighted mean for the item and
how would you interpret it? Would the answer be Less
Effective? Or perhaps Not Effective ?
To compute the weighted mean (Wm), we use the following formula:
Wm = the summation of the products of the respective
frequencies and weights divided by the number of cases. To
illustrate, we suppose that the results of tallies are as follows:
Category Frequency (f) Weight(x)
Very Effective (VE) - 83 5
Effective (E) - 51 4
Less Effective (LE) - 32 3
Not Effective (NE) - 24 2
Don't Know (DK) - 10 1
Total = 200
What could be the weighted mean ? We now take a look at its
computation.
 We compute the summation of the products of
the frequencies (f's) and weights (x's), that is,
Wm = [(83(5) + 51(4) + 32(3) + 24(2) +
10(1)] ÷ 200
Wm = (415 + 204 + 96 + 48 + 10) ÷ 200
Wm = 773 ÷ 200
Wm = 3.865. How do we interpret this ?
Is this VE ? or E ?
We then take a look at how to distribute the
weighted means across all the five (5)
categories.
 Let us compute an interval i. It can be
found by dividing the difference of the
highest weight (5) and the lowest weight
(1) by the number of options which is 5,
that is, (5 - 1) = 4 divided by 5 = 0.80. We
then distribute the weighted means across
all the five (5) categories using 0.80 as the
interval.
 Using 0.80, we determine the lower limits. We start
with 1.00 (bottom, pls see arrow) as the lower limit of the
first category. then the next category above 1.00 has a
lower limit of 1.80, that is, 1.00 + 0.80 and so on until
4.20, the first entry of the column. That is,
4.20 -
...
...
1.80 -
1.00 -
We then take a look at the upper limit.
To illustrate, in the interval 1.00-1.79, the lower limit is
1.00 and the upper limit is 1.79.
We then prepare to distribute the options such
as Very Effective (VE), Effective (E), Less
Effective (LE), Not Effective (NE), or Don't Know
(DK). We arrange them as follows:
Very Effective
Effective
Less Effective
Not Effective
Don't Know
 To recall, our lowest category among the
intervals is 1.00 - 1.79. Note that the gap is 0.79.
Then next upper category starts with 1.80. We
again add 0.79 to 1.80 that is, 1.80 + 0.79 =
2.59. Thus, a higher interval from below is seen
as 1.80 - 2.59, above the lowest. The process
continues till the highest interval which is 4.20 -
5.00, which completes the categories.
 And so we now have the categories are as follows:

4.20 - 5.00 Very Effective (VE)

3.40 - 4.19 Effective (E)
2.60 - 3.39 Less Effective (LE)
1.80 - 2.59 Not Effective (NE)
1.00 - 1.79 Don't Know (DK)
Hence, our weighted mean from a sample problem
above which is 3.865, implies that the policy is Effective
as seen within the second interval 3.40 - 4.19.
Practice Exercises: ( not to be submitted)

1. Give the distribution of weighted means

among the following set of options Very Satisfied
(VS), Satisfied (S), Moderately Satisfied (MS)
and Not Satisfied (NS), with 4 as the highest
weight and 1, the lowest. What is the
interpretation of a hypothetical mean of 2.06 ?
2. With 3 as the highest, how do you distribute
the weighted means with the following weights,3,
2, and 1? Use the categories Very Effective,
Effective and Not Effective.
 Activity 2: Research Statistics
( To be submitted; Due: June 14, 2025)

Give your solution for each of the following:

1. Show a distribution of weighted means for the following categories based
on 5 weights: Very Effective (VE), Effective (E), Somewhat Effective (SE),
Less Effective (LE), and Not Effective (NE). The weights are 5,4,3,2, and 1
respectively with 5 as the highest and 1, the lowest.
Under what category does a weighted mean of 2.76 belong ?

2. Determine what proportion (n) of a population N of 4,000 will make

your research work scientific in terms of sampling? Use the Slovin's
formula at 0.05 level of significance.
 Activity 3: For Research
( To be submitted; Due: June 14, 2025)

Read and analyse:

1. Instrumentation
2. Ranking in Statistics
3. Parametric tests
4. Analysis of Variance (ANOVA).

● Submit Activity 3 separately from

Activity 2.
References:

1. The International Statistical Institute, (2003). " The Oxford Dictionary of

Statistical Terms". edited by Yadolah Dodge, Oxford University Press.
2. International Encyclopedia of Statistical Science. (2014) "Sample Survey
Methods", edited by Miodrag Lovric. .
3.. Malec, M.A. (1993). Essential Statistics for Social Research. Colorado:
Westview Press, Inc.
4. Pino, A.P.Jr. (2010). Research Statistics. Cebu City: Krest Pub Hse.
5. Slavin, R.E. (1984). Research Methods in Education, A Practical Guide.
New Jersey: Prentice-Hall, Inc.
6. Verney,M.et.al. 2002. Behavioral Statistics in Action. 3rd Ed. Boston:
McGraw Hill Cos., Inc.
.