DATA ANALYSIS

WAN NORAZI BINTI MIOR ISMAIL

ADVENCE DIPLOMA IN PREVENTION AND INFECTION CONTROL
ILKKM SUNGAI BULOH
LEARNING OUTCOMES

At the end of the learning session, the

students should be able to:

▪Classify measurement scale in data analysis.

▪Explain the preparatory process for data
analysis.

DATA ANALYSIS 2
DATA ANALYSIS
Is the reduction and organization of a body of data
to procedure results that can be interpreted by the
researcher.
▪ A variety of quantitative and qualitative methods
may be used, depending upon the nature of the data
to be analyzed and the design of the study.
▪ For quantitative data, statistics are used to
interpret the finding

DATA ANALYSIS 3
Why do we analyze data?
The purpose of analyzing data is to obtain usable and
useful information. The analysis, irrespective of
whether the data is qualitative or quantitative, may:
▪ describe and summaries the data
▪ identify relationships between variables
▪ compare variables
▪ identify the difference between variables
▪ forecast outcomes

DATA ANALYSIS 4
 MEASUREMENT SCALES

▪ Refer to the classification of measurement according

to whether the obtained scores reflect a category
(quality) or a numerical value (quantitative).
▪ 4 common measurement scales:

DATA ANALYSIS 5
 MEASUREMENT SCALES

CATEGORICAL
▪ Each category possesses certain characteristics and
individuals are categorized by their possession of
those characteristics that defines a subset.
▪ Example: marital status, gender, nationality.

DATA ANALYSIS 6
MEASUREMENT SCALES
Nominal scale
▪ A nominal scale is where: the data can be classified
into non-numerical or named categories, and the
order in which these categories can be written or
asked is arbitrary.
▪ Nominal scales could simply be called “labels.”
▪ Example: Gender (1= male, 2= female), hair color,
marital status, religious, blood type.

DATA ANALYSIS 7
DATA ANALYSIS 8
DATA ANALYSIS 9
Ordinal scale
▪ An ordinal scale is where: the data can be classified
into non-numerical or named categories an inherent
order exists among the response categories.
▪ Ordinal scales are seen in questions that call for
ratings of quality (for example, very good, good, fair,
poor, and very poor) and agreement (for example,
strongly agrees, agree, disagree, strongly disagree).

DATA ANALYSIS 10
Examples of Ordinal Scales

DATA ANALYSIS 11
DATA ANALYSIS 12
 MEASUREMENT SCALES

CONTINUOUS
▪A continuous variable is a variable that has a
range of values that is infinite but divisible.
▪Example: income, age, temperature, tests
scores.

DATA ANALYSIS 13
 MEASUREMENT SCALES

Interval
▪ Show ranking of events or objects on a scale with
equal intervals between the numbers.
▪ The interval scale possesses, two out of three
important requirement of a good measurement scale,
i.e. magnitude and equal intervals, but lacks the real
absolute zero point.
▪ Interval data has order and equal intervals.
▪ Measured on a linear scale and can take positive and
negative values.
▪ Example: temperature of 40˚C is 10˚C higher than 30˚C
and an increase from 20˚C to 40˚C is twice as much as
the increase from 30˚C to 40˚C.
DATA ANALYSIS 14
 MEASUREMENT SCALES

Ratio Measurement
▪ Ratio level measurements have all three attributes
magnitude, equal intervals and absolute zero point.
Thus, magnitude from zero to some point on the
scale is known.
▪ Ratio measurement is very similar to interval
measurement except that ratio measurement
involves the use of a true absolute zero.
▪ Example: Height, weight, pulse and blood pressure. A
zero on a ruler means no length and 20 cm is twice
as long as 10 cm.

DATA ANALYSIS 15
 FOUR LEVELS OF MEASUREMENT OF
DATA

DATA

QUALITATIVE / QUANTITATIVE /
CATEGORICAL NUMERICAL
(can be grouped) (can measured)

NOMINAL ORDINAL INTERVAL RATIO

cannot be arranged in Can be arranged in order Zero is arbitrary Zero is arbitrary and
any particular order means none
BEST MEASURE OF BEST MEASURE OF
BEST MEASURE OF CENTRAL TENDENCY: CENTRAL TENDENCY: BEST MEASURE OF
CENTRAL TENDENCY: MEDIAN, MODE MEAN, MEDIAN, MODE CENTRAL TENDENCY:
MODE MEAN, MEDIAN, MODE
BEST MEASURE OF BEST MEASURE OF
BEST MEASURE OF SPREAD: IQR SPREAD: RANGE, BEST MEASURE OF
SPREAD: NONE VARIANCE, SD AND IQR SPREAD: RANGE,
VARIANCE, SD AND IQR

DATA ANALYSIS 16
DATA PROCESSING
DATA PREPARATION,
INPUT, VERIFICATION
AND CLEANING

DATA DATA
CLEANING EDITING

DATA DATA
VERIFICATION CODING

DATA
ENTRY
DATA ANALYSIS 17
1. Data Preparation, Input,
Verification and Cleaning

▪Data need to transfer from the data

collection tool (e.g. checklist, rating scale
and questionnaire) into format that suitable
for analysis.
▪Before analysis take place, a researcher
should check the accuracy of the data to
deal constructively with any missing values.

DATA ANALYSIS 18
2. Data Editing
▪ This is scrutinizing process of data and occur almost
every phase of data collection or analysis
▪ It ranges from the almost routine activities of
correcting typographical errors to out range entries
done by researcher.

DATA ANALYSIS 19
3. Data Coding
▪ Coding is a process of assigning numerals or other
symbols (character codes) to classify answers of all
responses so that responses can be put into a
limited number of categories or classes. This
enables us to enter data for analysis.
▪ Coding is the process by which the original data are
transformed into symbols compatible with
computer-assisted analysis or other types of
analysis.

DATA ANALYSIS 20
4. Data Entry
▪ Data entry is a process to enter data code into a data
sheet.
▪ A Researcher can directly enter the data code into a
data analysis package (such as SPSS) or into a spread
sheet (such as MS excel).
▪ Direct imputation of data SPSS or excel spread sheet
allow speedy preparation of frequency tables, figures
and charts.
▪ However, online questionnaire provides an option to
import data directly into a software package, thus
saving data input time.

DATA ANALYSIS 21
5. Data Verification
▪ Once the data have been entered into computer
package, they need to be checked carefully for accuracy.
▪ Verification is an attempt to ensure error should not
occur when transferring data from one source to
another.
▪ A random examination of the coded data also makes
sure to find out discrepancies.

DATA ANALYSIS 22
6. Data Cleaning
▪ Data cleaning aims to identify and correct the error or a
least to minimize their impact on study results.
▪ Data cleaning deal with data problem once they occurred
▪ Data cleaning involve two types of checks.
▪ The first check is for outlier and wild codes.
▪ Outliers are value that lie outside the normal range, it can
be found by inspecting frequency distribution (lowest &
highest values).
▪ Another problem is a wild code. For example, a researcher
assigns code for gender; (1) Male and (2) Female. If some
of gender found with code 3, is the wild code.

DATA ANALYSIS 23
HANDLING ERRORS IN DATA
1. Blank Responses
▪ One problem faced when collecting data is that some
of the data may be missing. For example, in
conducting a survey with ten questions, perhaps some
of the people don’t answer all ten questions.
▪ A simple approach for dealing with missing data is to
throw out all data for any sample missing one or more
data elements.
▪ One problem with this approach is that sample size will
be reduced.

DATA ANALYSIS 24
HANDLING ERRORS IN DATA
2. Outliers
▪ Deleting the values: The outliers can deleted if the
outliers are wrong or if the reason the outlier was
created is never going to happen in the future.
▪ For example, there is a data set of peoples ages and
the usual ages lie between 0 to 90 but there is data
entry off the age 150 which is nearly impossible. So,
we can safely drop the value that is 150.
▪ Using different analysis methods: Researcher could
also use different statistical tests that are not as much
impacted by the presence of outliers.

DATA ANALYSIS 25
DATA ANALYSIS 26