CHAPTER NINE – DATA ANALYSIS IN QUANTITATIVE AND QUALITATIVE

RESEARCH METHODS

Learning Objectives:

At the end of this module, the students will be able to:

◦ discuss the value of knowing how to analyze both quantitative and qualitative data;
and
◦ show thorough understanding of the process of data analyses by giving correct
computations in quantitative data or illustrations asked for in the problem.

DATA ANALYSIS IN QUANTITATIVE AND QUALITATIVE RESEARCH METHODS

The concept of analyzing research data is greatly dependent on the research design
utilized in the study; as well as on the types of questions that were asked in order to answer
the problems of the study; and the research hypothesis that was advanced.
Research data may be either quantitative (numerical) or quantitative (verbal) or both
numerical and verbal in the case of the mixed-method design. When the data are already
available, they have to be sorted out and assessed according to the types of questions
asked in the problem statement of the study before they can analyzed and interpreted. It is
through the process of data analysis that the researcher is able to make a clear description
about the responses of the research participants, noting what is typical or extraordinary
about the data, observing difference or relationships, and finding answers to the research
questions or hypothesis.

QUANTITATIVE DATA ANALYSIS

Several authors sound off the idea that research is a scientific process to verify ideas
and theories by gathering information in order to answer the research questions. Statistical
analysis is relevant for research projects wherein numbers represent the information
gathered. The numerical information gathered to describe age, income, or levels of
agreement or disagreement are referred to as data. According to Healey (2010), statistics
are mathematical techniques used to examine data in order to answer questions and test
theories. For any researcher, data can be most trustworthy information which, if properly
collected and thoroughly analyzed, become the most objective foundation about the topic of
research. Statistics must be applied effectively to organize, evaluate, and analyze the
research data in order to come up with meaningful information that serves to answer the
research problems. The two classes of statistical techniques are descriptive statistics and
inferential statistic.
DESCRIPTIVE STATISTICS
According to Coakes (2013), descriptive statistics are used to explore the data
collected and to summarize and describe those data.
Research results may be represented in terms of percentages, proportions, ratios
and rates.
The following statement which is a research finding may be considered:
There are 300 students in the school who were surveyed regarding their opinions
about the effects of prohibited drugs. Of this number, 200 said that prohibited drugs are
dangerous; 55 said that drugs have some good effects in the body; while 45 said they had
no idea. There seems to be nothing wrong with the statement, but the same finding could be
better conveyed in terms of percentages; i.e., almost 67% of the 300 total respondents said
that prohibited drugs are dangerous; 18% said that drugs have some good effects in the
body; while 15% said they have no idea.
As pointed by Healey (2010), percentage and proportion provide a frame of reference
for reporting research results by standardizing the raw data, that is, percentage to the base
100 and proportion to the base 1.00. the mathematical definitions of percentage and
proportion are:
Proportion : P = f/n
Percentage : % = (f/n) x 100
Where:
f = frequency or the number of cases in any category
n = the number of cases in all categories

Table A
Opinions of 300 Students on the Effects of Prohibited Drugs

Opinion of Respondents Frequency Proportion Percentage

(f) (p) (%)

1. Prohibited drugs are 198 0.66 66%

dangerous to the health of a
person

2. Drugs have some good 60 0.20 20%

effects in the body

3. No idea 42 0.14 14%

Total 300 1.00 100%

 Research results may also be summarized simply and clearly through ratios and
rates. Ratios are useful for comparing the number of cases in the categories of a variable.
In this statistic, the frequency of one category is divided by the frequency in another category,
instead of standardizing the distribution of the variable to the base 100 or 1.00 as what is
done in computing percentages and proportions.
Below is the mathematical definition of a ratio:
Ratio = f1/f2
where: f1 = the number of cases in the first category
f2 = the number of cases in the second category

To illustrate:
Suppose the researcher wants to know the relative sizes of the various elementary
schools in a district, and he found that there are 130 public elementary schools and 85
private elementary schools. To determine the ratio of the public schools (f1) to private
schools (f2), we divide 130 by 85, thus:
Ratio = f1/f2 = 130/185 = 1.529 or 1.53
The computation shows that the ratio is 1.53 which means that for every private
elementary school, there are 1.53 public schools in the district covered by the study. This
result can be expressed thus: based on a unit or ones, the ratio of public elementary schools
to private elementary schools is 1.53:1
On the other hand, rates provide yet another way of summarizing the distribution of a
single variable. Rates are defined as “the number of actual occurrences of some
phenomenon divided by the number of possible occurrences per some unit of time” (Healey,
2010). For example, the percentage of passing rate of board passers in a university refers
to the number of passers in that university (actual occurrences) divide by the number of
takers or graduates who took the board exam (possible occurrence). This quantity is then
multiplied by 100.
Formula:
Percentage of Passing Rate = number of passers (95) X 100
number of takers (120)
= 95/120=0.7916 x 100
= 79.16%
Thus, based on the results of these computations, it is observed that the subject
university obtained a 79.16% rate of passing in the board examination.
 Another technique of presenting large sets of data is through the frequency
distribution. As discussed by Johnson and Kuby (2013), a frequency distribution is a listing,
often expressed in chart form, that pairs values of a variable with their frequency, or the
number of times the value (x) occurs in the sample. For instance, a study concerns the
number of females with varying ages in a group of twenty contestants in a song contest.
The data may be presented as follows:
Table B
Frequency Distribution of 20 Female Contestants According to Age

Age (X) Frequency (f)

10 - 1
15 - 2
17 - 5
20 - 7
25 - 2
27 - 2
30 - 1
Total = 20

At a glance, the data in Table B gives a picture of the number of female contestants
in every age category. However, when dealing with a larger set of data, the grouped
frequency distribution is more practical. For example, the research is about determining the
number of male evacuees in a relocation center, categorized according to their ages. Table
C is shown as a sample.
Table C
Grouped of Frequency Distribution of 100 Males in a Relocation Center

Age (X) - Frequency (f)

0-5 - 20

6-11 - 15

12-17 - 15

18-23 - 10

24-29 - 5

30-35 - 3
 36-41 - 3

42-47 - 2

48-53 - 3

54-59 - 10

60 and above - 14

Total = 100

In this grouped frequency distribution, the arrangement of data makes it easier to

detect how many males have been relocated according to their age groupings.

MEASURE OF CENTRAL TENDENCY

According to Johnson and Kuby (2013), these are numerical values that locate, in
some sense, the center of a set data. Measures of central tendency are expressed as the
mean, median, mode and midrange.
1. The Mean, which is also called arithmetic mean, is represented by X (or “x-bar). This is
computed by adding all the values of the variable x, (The sum of the x values is symbolized
by x or summation of x) and dividing the sum by the total number of samples, represented
by n. The formula for this is expressed as:

Example = X = x = 10+15+17+20+25+27+30 = 144 = 20.57

n 7 7
The example shows the ages of seven students enrolled in an alternative class. After
computing the mean value, it is determined that the average age of the seven students in
that class is 20.57. Another description for the mean value is that it represents the middle
point of a group of values according to weight.
2. The Median. When the date are ranked in proper order according to size, the value of the
data that occupies the middle position is the median. The sample median is represented by
x (or “x-tilde”) or sample median. To find the median, three steps have to be followed: (1)
rank the data (2) determine the depth of the median (or number of positions from either end):
depth of median d (x) = sample size +1 divided by 2. In our given example, the depth of the
media is d(x) = 4. (3) Determine the value of the median by counting the ranked data,
locating the data in the d ( x ) position.
To illustrate, the median of the given data, (10,15,17,20,25,27,30) is the fourth value from
the lowest rank and likewise from the highest rank. Therefore, in this case the sample
median age is 20(x=20).
3. The Mode. This refers to the value of x that occurs most frequently in a set of data. If
there is no number that occurs more than once, the sample has no mode.
4. Midrange. This refers to the number which is found exactly midway between a lowest
data value (L) and a highest data value (H). To find the midrange, get the average of the low
and high value. In our given example, the lowest value is 10 and the highest value is 30,
thus L=10 and H=30
= 10 + 30 = 40 = 20. The midrange is 20.
2 2

The values representing the four measures of central tendency may be similar, but
more likely they will be different. However, the four values can be used to describe the
center in a set of data.

MEASURES OF DISPERSION
Another way of analyzing quantitative data is to look at the spread, or variability of
the set of data, and we are referring to the measures of dispersion, which include the range,
variance, and standard deviation. The simplest measure of dispersion is the range, which is
the difference in value between the highest value (H) and the lowest value (L) in a set of
data. For example: 5,4,6,7,9 has a range of H-L = 9-4 = 5. The range of 5 means that these
data all fall within a 5-unit interval. In computing the sample variance or S2, we need to
understand the deviation from the mean which simply refers to the difference between the
value of x and the mean ,i.e. (x- x). When the deviation from the mean are squared and
added, the formula for sample variance can be applied as:
S2 = Σ (x- x)2
n-1
Inferential Statistics
This type of statistics makes use of the information contained in the sample data to
increase the researcher’s knowledge about the sampled population. Through the Sampling
Distribution of Sample Means (SDSM), two types of inferences can be made: 1. estimating
the value of a population parameter and 2. testing a hypothesis, (Johnson & Kuby, 2013).
A hypothesis is a tentative statement that something is true, which needs to be
proven through statistical information. In research, the researcher pursues a belief or
concern that an idea is true; however, for the purpose of objectivity, a scientific testing
process has to be done. The null hypothesis, express as Ho, is the hypothesis subjected to
testing. The alternative hypothesis, written as H1, is also known as the research hypothesis
since it represents the idea that the researcher theorizes in his study.
An example for the t-test of difference is shown below:
Ho : There is no significant difference between the math scores obtained by
students enrolled in the science section and those enrolled in the regular section.
H1 : Students enrolled in the science section will have higher math scores than
those enrolled in the regular section.
Decision Rule : (Johnson & Kuby, 2013)
a. If the p-value is less that or equal to the level of significance (α≤.05), the decision
must be to reject Ho.
Results : calculated p-value = 0.035
Decision about Ho : Reject Ho.

Conclusion about H1 : There is a probability at the .05 level of significance that the
students under the science section will have higher math scores than those enrolled in the
regular section.
A study which calls for a measurement of relationship may utilize the statistic of
correlation which looks at the relationship between two variables in a linear fashion. The
Pearson product-moment correlation coefficient or Pearson’s r describes the relationship
between two continuous variables (Coakes, 2013). A hypothesis which calls for a test of
relationship may be presented and treated in a similar pattern as the test for differences;
however, the Pearson’s r is the statistic to be computed.

Qualitative Data Analysis

Qualitative data represent the information gathered in qualitative researches. These
are data that come in words rather than numbers. The concern in qualitative research is on
meanings; hence context and experience are treated as important. The researcher values
subjectivity, and qualitative data are analyzed backed by philosophical assumptions which
could be ontological, epistemological, or axiological assumptions.
Qualitative research involves thinking qualitatively. The purpose of qualitative
research is to understand or explore meaning and the ways people make meaning, rather
than to prove a theory or determine a relationship between factors” (Braun & Clarke, 2013).
The analysis of qualitative date could involve several approaches such as discourse
analysis, content analysis, thematic analysis, biographical or narrative analysis and
grounded theory.