Data Processing and

Presentation
JOYCEE D. LOQUITE
Discussant/Reporter
Contents and overview
• Data processing activities
• Frequency distribution table
• Data reduction and coding
• Kinds of codes
• Questions for self evaluation
Data processing Activities

Data processing activities consume much of the time of

researcher after the data are gathered from the field. Data
processing therefore refers to the process of transforming raw
data into meaningful output i.e. information. It can be done
manually using pen and paper. Mechanically using simple
devices like typewriters or electronically using modern data
processing tools such as computers.
The frequency Distribution Table
•What is a Frequency Distribution Table?
•Frequency tells you how often something happened. The frequency of an observation tells you
the number of times the observation occurs in the data. For example, in the following list of
numbers, the frequency of the number 9 is 5 because it occurs 5 times.

1, 2, 3, 4, 6, 9, 9, 8, 5, 1, 1, 9, 9, 0, 6, 9.
   
Method Number
Abstinence 14
Condoms 47 A frequency distribution table showing categorical variables.
Injectables 1
Norplant 1 Frequency distribution tables give you a snapshot of the data to
Pill 35
None 307 allow you to find patterns. A quick look at the above frequency
Total
 
405
distribution table tells you the majority of teens don’t use any
birth control at all.
Frequency Distribution - Example
We had 183 students fill out a questionnaire. One of the questions was which study
major they're following. The screenshot below shows part of these data.
Frequency Distribution - Table
So what about these study majors? Just gazing at our 183 values isn't going to help us. A
more viable approach is to simply tabulate each distinct study major in our data and its
frequency -the number of times it occurs.
The resulting table (below) shows how frequencies are distributed over values -study
majors in this example- and hence is a frequency distribution.
Data Reduction and Coding
Data reduction refers to the process of selecting, focusing, simplifying, abstracting, and transforming the data that
appear in written up field notes or transcriptions." Not only do the data need to be condensed for the sake of
manageability, they also have to be transformed so they can be made intelligible in terms of the issues being
addressed.
Data reduction often forces choices about which aspects of the assembled data should be emphasized, minimized, or
set aside completely for the purposes of the project at hand. Beginners often fail to understand that even at this stage,
the data do not speak for themselves. A common mistake many people make in quantitative as well as qualitative
analysis, in a vain effort to remain "perfectly objective," is to present a large volume of unassimilated and
uncategorized data for the reader's consumption.

Coding of data
Coding of data refers to the process of transforming collected information or observations to a set
of meaningful, cohesive categories. It is a process of summarizing and re-presenting data in order
to provide a systematic account of the recorded or observed phenomenon. Data refer to a wide
range of empirical objects such as historical documents, newspaper articles, TV programming,
field notes, interview or focus group transcripts, pictures, face-to-face conversations, social media
messages (e.g., tweets or YouTube comments), and so on. Codes are concepts that link data with
theory. They can either be predefined by the researcher or emerge inductively from the coding
process. 
Scales of Measurements

ARLENE D. DERUTA
Discussant/Reporter
Contents and Overview
• Scope of Measurement
• Definition of Measurement
• Types of Measurement
• Nominal Scale
• Ordinal Scale
• Interval Scale
• Ratio Scale
Measurement Defined

 Measurement is the assignment of number or symbol to an event, an object or an observation based on logically accepted rules.

 Is the process of observing and recording the observations that are collected as part of a research effort.
 

Types of Measurement

1. Constant Measurement ---the observation does not vary, thus the term constant. In other words, constant
measurement assumes only one single value.
 
2. Variable Measurement --- varies or may assume two or more values along a continuum.
 
Variable Measurement may also be categorized into
1. Qualitative (varies in kind)
Example: sex can be male or female
Religion can be catholic on non-Catholic
2. Quantitative (varies in degree or magnitude)
Example: Income, the values or categories are given in terms of quantity
Such as 5, 000 per week or 10, 000 per month
 
 Four kinds of scales
 
1. Nominal Scale

This scale is also called classificatory scale and is considered as the weakest scale of
measurement. In here, the data can be sorted into categories that are mutually exclusive which
are then assigned symbols or labels. For example, a group of students may be sorted into
various categories as “freshmen, sophomores, juniors and seniors or may be classified
according to courses they are taking such as education, engineering, computer, tourism and
others.
 
Example in this case is religion: say Catholics, INC, JIL and Aglipayan. This religion can be
assigned as a number 1, 2, 3, and 4 respectively, but none is greater than the other.

It is important to note that in the nominal scale, equivalence is the only property that defines
relationship among members of each category. In this case, the members of a set assigned to
the same class or category are equal in the characteristics being scaled. This property of
equivalence exists not only among members within each class but among several classes as
well.
2. Ordinal Scale

Is a measurement scale characterized by the presence of categories which are

classified according to order: ranking is often the term used. Using this scale, the
data are not sorted only but are rank as well, hence the name ranking scale. This
scale is used whenever we are interested to indicate a greater than or a less than
relationship among or between objects or individuals or whenever we use a rating
scale such as 0, 1, 2 etc.
 
Example of data in the ordinal scale include juiciness of hotdogs, tenderness of
meat, quality of construction, hardness of minerals, attitudes and emotions, socio-
economic status, and the like. Note that in this scale, data can be ranked from most
to least or vice-versa.
3. Interval Scale

The interval scale is considered as the first truly quantitative scale. It has all the
characteristics of an ordinal scale and the distance or difference between any two
numbers on the scale. This scale thus provides numbers which reflect differences
among items. In an interval scale, the difference between any two numbers on the
scales is of known size and the categories are of equal interval. The numerically
distances represent equal distances in the properly being measured.
 
Most of the measurements done in the field of sociology, psychology and
education make use of the interval scale. Statistical measures like the arithmetic
mean, standard deviation and product moment correlation coefficient may be
computed from interval data. In fact, practically all the common parametric
statistics are applicable to interval data.
4. Ratio Scale

The highest scale of measurement is the ratio scale. This has all the characteristics
of the interval scale including an absolute zero (true zero point as its origin). A
zero measurement on the ratio scale, implies an absence of a particular attribute or
characteristics of the property being measured. Since there is true zero point all
arithmetic operations are possible in this scale. This means that the operations of
addition, subtraction, multiplication and division can be done in the ratio scale.
The ratio scale can be used with anything that can be counted, weighed or
Measured in exact amount.
 
Example of ratio variable or data includes income, age, count of people, years of
schooling, work experience and the like. All ratio data therefore are amenable to
parametric statistical analyses.
Basic Statistical Tools for Data
Analyses

ROGEN MAE D. DIONIO

Discussant/Reporter
 CONTENTS and OVERVIEW
 Roles of Statistics in Research
 Qualitative Data Analysis
 Quantitative Data Analysis
 The Normal Distribution
 Measures of Central Tendency
 Considerations in the Use of Averages
 Measures of Dispersion or Variability
 The Range
 The Variance
 The Standard Deviation
 Pearson Product Moment Correlation Coefficient
 Spearman’s Rank Order Correlation Coefficient
 Chi- Square Test of Association
 Test of Significance of Mean Differences
 Interpretation of Statistical Results
Statistics- is the process concerned with the
collection, consolidation, analyses,
and interpretation of research data.
Role of Statistics in Research
1. Normally use to analyze data.
2. To organize and make sense out of large amount of data
3. This is basic to intelligent reading in research article
4. Has significant contributions in social sciences and even business and
economics
5. Statistical researchers make inferences about population characteristics
on the basis of one or more samples that have been studied.
Types of Data Analysis
1. Qualitative Data Analysis
 This is non-quantitative technique of analyzing research data.
 The researcher interprets the data without putting emphasis on qualifications but deals
largely on quality information.
 Through a more in-depth and analytical examination of data, new insights, ideas, and
may be trends could be drawn where quantitative technique may not apply.

2. Quantitative Data Analysis

 Quantitative Analysis It deals with the data expressed in quantitative form or in numerical.
This technique essentially employs statistical tests or procedures as means of presenting and
interpreting data.

• Population is the entirety of the including all the members that forms a set of data.
• Sample- contains a few members of population. Samples were taken to represent the characteristics
or traits of the population.
 Two types of statistical analyses
1. Descriptive statistics
- used to say something or describe a set of information collected. It can also
be presented with graphs.

2. Inferential statistics
-Use to say something about a larger group (population) using information
collected from a small part of that population (sample).
-Taking samples to represent your population
 
 The Normal Distribution
A Normal Distribution or normal curve- is the distribution
which characterizes many of our research data and information.
It is an important concept which has many practical
applications in the field of statistics, psychology, biology,
education, and other sciences.

The normal curve is a bell- shaped with its central point being
the highest point in the curve. This central point is where the
mean is located. Note that the distributions or scores tend to
cluster. The curve is also symmetrical around its vertical axis.
Figure 11: Percentage under the normal curve at various
standard deviation units from the mean
Measures of Central Tendency
The measure of central tendency most commonly
known as average numerical values which locate the
center or middle of a set of data. The most common
average measures are the mean, median and mode.
Each gives a different picture of the data it is called for
to represent.
 The Arithmetic Mean
The arithmetic mean or simply the mean is the most commonly
used measure of tendency. It is obtained by adding the items in the
series, then dividing their sum by the number of items added. This can
be stated in terms of formula as follows:
Mean = “X
N
Where: = is the Greek symbol, sigma, which
means “take the sum of” ….”
X = the individual item, score, or observation
N = total number of items, score, or
observation.
For grouped data when the scores or observation have been entered into a
frequency table, the mean is computed using the formula:
Mean = FX/N
Where = Notation for “take the sum of”
X= class mark or midpoint of a given class interval
F= Frequency corresponding to a given class
interval
N= Total number of cases or observations
The median
Median is the middle value on an array.
Array is an arrangement of a set of numbers or items of
observations according to their order of magnitude (either increasing or
decreasing).
For example, let us take the first five incomes of farmers in the first
column as a set of data. And array for this set is P 1,766, P1,862, P
1,874, P 1,950, and P 2,700 (or the reverse). The income P1,874 is the
third or middle value in this array and so the median is P 1, 874.
The Mode
The mode of a set of number of observations is that the value which
occurs with the greatest frequency, i.e., it is the most common value.
Considerations in the Use of Averages

Two important factors must be considered in the choice

of which measure of average is going to be used.
1. Appropriateness of the measure to the purpose
2. The type of data to summarize
x Income (P) X Income (P)

1 2700 6 1862

2 1950 7 2100

3 1874 8 1999

4 1862 9 1974

5 1766 10 1974
•The
  Standard Deviation
The standard deviation of a set of N numbers is the positive square root of the
arithmetic mean of the squares of the deviations from the mean. More aptly, it is the
square root of the variance. When it is necessary to distinguish the standard
deviation of a population from that of sample drawn from this population, the
symbol is s used for the latter and a, for the former. Thus, s would represent the
sample and population standard deviation, respectively. Since it is the square root of
the variance, the standard deviation for the sample data on the income of 10 farmers
is 260.81 which means that the farmers’ income deviates with this value from the
mean.
s=
s=
 
Uses of the Different Measures of Dispersion
Range – as a measure dispersion is easily obtained and understood.
However, it is very unreliable since its computation is based only on two
values, the lowest and highest value in the distribution.
Variance- is a characteristic which satisfies the characteristics which
satisfies the criteria of a good measure of variability. Namely:
1) It is computed from all the X-values in the distribution.
2) it is independent of the number of cases.
3) it takes into account the measures of central tendency.
A higher value of variance indicates higher variability between the
two groups, i.e., the more the two groups are different on a particular
characteristic being studied.
Measures of Relationship and Association
What we have presented so far are measures which we call descriptive
statistics. They are used whenever we want to draw a numerical picture or
describe a particular population. I f our interest is to find out whether a
relationship or an association exists between two or more variables,
however, then we can select from a number of tests we call measures of
association, a specific one suited to the level of measurement of our
research data.
Pearson Product Moment Correlation Coefficient
The Pearson product-moment correlation has been used by researchers to compare time series
data sets to assess the temporal similarities. Computer generated data, vertical ground reaction
force (VGRF) data and hybrid data (constructed by combining features of computer generated and
VGRF data) were used to investigate the influence of timing and amplitude differences on the
effectiveness of this technique.
There are three approaches to interpret a correlation coefficient r. This interpretation is dependent
upon the researcher’s purpose and the situations that may influence the size of the correlation. The
most common approach to interpret r is giving a qualitative description to the size of r as follows.
Coefficient (r) Description

.00 to 20 little or no relationship

.20 to .40 some slight relationship

.40 to .60 substantial relationship

.60 to .80 strong useful relationship

.80 to 1.00 high to very high relationship

Spearman’s Rank Order Correlation Coefficient
This measure or method is a form of the Pearson product moment correlation most suited to
situations where the number of cases is 25 to 30 or less. The symbol used to represent this
correlation coefficient is p (rho). Specifically, rho is only applicable to paired variables which
are ranked. It uses differences between ranks to compute a statistics that is then evaluated by
means of a table determine significance.
Spearman’s rho is much easier and quicker to compute than Pearson’s r. It may be interpreted
in the same manner as Pearson’s r. To compute this coefficient the following formula is used:
P= 1 - 6∑D2
N (N2-1)

Where:
D= the difference between paired ranks
D= the sum of the squared differences between ranks
N= number of paired ranks
 
Chi -Square Test Association
 
The Chi- square test, symbolized as X, is one of
the simplest yet most useful among statistical
tests.
Test of Significance of Mean Differences
The mean is described to be the most satisfactory measure in
characterizing a group or population. Hence, differences between
means of samples need to be evaluated to determine whether they
are significant.
The Analysis of Variance
 
Popularly called by its acronym, ANOVA or ANOV, the
analysis of variance is an effective method of determining
significant differences among means of more than two groups or
populations. The ANOVA technique thus is so designed such that
all of the data are treated at once and a general null hypothesis of
no difference among the means of the various groups is tested.
The technique involves statistical computation of the variance of each sample.
This introduces us to three types of variances:

Total-group variance is the variance of the scores for all the groups combined
into one composite group.

Within-groups variance is the average variability of the scores within each group
The within-group variance represents the sampling error in the distribution and is
also referred to as error variance

Between-group variance is the variation of group means from the total grand
mean of all groups. It represents the influence of the experimental variable of the
dependent variable.
Interpretation of Statistical Results
 

The interpretation of statistical results is based on internationally accepted rules. On this basis, the

results may be declared as not significant or statistically significant, depending on the decision arrived at

as revealed by the statistical analysis. The statistical decision is either to reject or not the null hypothesis.

This decision is usually based on the relative values of the computed and tabular F-values. The statistical

interpretations are described as follows:

Not significant – when the null hypothesis is not rejected, that is the computed F-value is lower than the

desired tabular value at a certain significant level, the result is said to be not significant (ns).

Statistically significant – The result is declared statistically significant when the computed F-value is equal

or larger than the tabular value at a desired level of significance. If the former is higher than the latter value

at 5% and 1% level, the result is considered significant and highly significant respectively. This suggests that

difference between a pair or more pairs of group means is real and not merely due to chance.