QUANTITATIVE DATA ANALYSIS AND

SUMMARIZATION
(MCPEWM 510)

LECTURER: MR.T. C. BHANYA

CELL: +263713354403/+263773937287
EMAIL: tbhanya@gmail.com
DATA ANALYSIS AND SUMMARIZATION

Applied Statistics

Descriptive Statistics Inferential Statistics

Estimation
Summary Graphical Hypothesis
of
Measures Displays Testing
parameters
 QUANTITATIVE DATA ANALYSIS

• Quantitative data refers to all data that can be reduced to numerical values, ranging from

the numerical frequency of occurrences to complex presentation of data in terms of

graphs and charts.

• To convey meaning within the framework of the study, this data has to be analyzed and

interpreted.

• In short, collecting and analyzing data with quantitative techniques includes understanding

the relationship among variables utilising descriptive and inferential statistics.

 LEVELS OF MEASUREMENT

• Quantitative variables take several forms, frequently called levels of measurement,

which affect the type of data analysis that is appropriate.

• Although the terminology used by different analysts is not uniform, a common way of
classifying a quantitative variable is according to whether it is nominal, ordinal,
interval, or ratio.
 DATA ANALYSIS PROCESS

Use
Descriptive Check
Categorisat Entering
Coding & Reliability
ion the data
Inferential & Validity
Statistics
 DATA ANALYSIS PROCESS - BRANCHES OF STATISTICS

Descriptive Statistics

• This is a branch that deals with different ways through which a given board of data can
be summarised or organised and presented

• It also presents information in a convenient, usable and understandable form

Inferential Statistics

• This is a branch that deals with drawing generalisations or conclusions about

populations using data drawn from samples
 WAYS OF SUMMARISING DATA

Summary Measures

• This is a way of condensing or summarising data through the computation of numeric

values, a single value will represent a particular data set of numeric measurements
i. Measures of central tendency (single statistical value that represents an entire distribution)
ii. Measures of variation/spread/dispersion (the spread of data)
iii. Measures of skewness (the degree of asymmetry of a distribution)
iv. Measures of position (determines the position of a single value [quartiles or percentiles])
v. Measures of kurtosis (tailedness of a distribution).
 WAYS OF SUMMARISING DATA CONT

Graphs

• These are visual displays which are used to represent data

• Pie charts; pictograms, bar graphs, line graphs

• Frequency histograms, ogive curves, stem and leaf, box and whisker plots, Lorenz
curve, scatter plots etc
 DATA ANALYSIS PROCESS

Frequency distributions
• The number of times various sub categories of a phenomenon occur, from which the
percentage and the cumulative percentage of any occurrence can be calculated.

Measures of central tendency

• There are three ways of measuring central tendency:
i. value that occurs the most frequently (mode);
ii. middle value or mid-point value after the data have been ranked (median); and
iii. value, often know as the average, that includes all data values in its calculation
(mean).
 DATA ANALYSIS PROCESS

Dispersion
• It is important to describe how the data values are dispersed around the central
tendency. Two of the most frequently used ways of describing the dispersion are the:
i. Difference between the lowest and highest values, that is, the range; and
ii. The extent to which the values differ from the mean, that is standard deviation
(Saunders et al., 2003).
 DATA ANALYSIS PROCESS - INFERENTIAL STATISTICS

While descriptive statistics are used to describe a sample’s characteristics, inferential

statistics are used to infer something about the population from which the sample was
drawn based on the characteristics of the sample.

i. t-Test - The t-test is a commonly used inferential test of the significance of the
difference between two means based on two independent, unrelated groups. Males
and females can be the two different groups
 DATA ANALYSIS PROCESS - INFERENTIAL STATISTICS

ii. The analysis of variance (ANOVA) - It is a statistical technique for examining the
differences among means for two or more groups.

iii. Chi-square analysis - Chi-square is used to make inferences when the data can be
divided into different categories. A chi-square analysis involves measuring participants in
terms of categories male-female, voter-non-voter and so on. By using the chi-square, the
researcher can determine if, for example, employees (male versus female) have their
own perceptions of their immediate supervisors.
 DATA ANALYSIS PROCESS - INFERENTIAL STATISTICS

iv. Correlation analysis - Correlations are used to describe relationships between

variables such as income, age and so on. For example, the size of a city and its crime
rate may be described with correlation analysis. Correlations estimate the extent to
which the changes in one variable are associated with changes in the other variable
(Welman et al., 2005). Pearson correlation coefficient is usually used to test the
relationship for interval and ratio scaled variables. The Spearman rank or the Kendall’s
Tau co-efficients are appropriate when variables are measured on an ordinal scale.
 DATA ANALYSIS PROCESS - INFERENTIAL STATISTICS

v. Regression analysis - The regression co-efficient is used to assess the strength of

relation between a quantifiable dependent variable and one or more quantifiable
dependent variables. The process of calculating a regression co-efficient using one
independent variable is called regression analysis. Calculating the regression co-efficient
using two or more independent variables is termed multiple regression analysis.
 DATA ANALYSIS PROCESS - RELIABILITY

Reliability
The reliability of a measure is established by testing for both consistency and stability. Two tests
of stability are test-retest reliability and parallel-form reliability.
Test-retest reliability - The reliability co-efficient is obtained with a repetition of the same
measure on a second occasion is called test-retest reliability.
Parallel- form reliability - When responses on two comparable sets of measures tapping the
same construct are highly correlated, that is, parallel-form reliability.
Consistency can be examined through the inter-item consistency reliability (the consistency of
respondents’ answers – Cronbach’s alpha) and split half reliability tests (correlations between two
halves of an instrument).
 DATA ANALYSIS PROCESS - VALIDITY

Validity
• Validity refers to the degree to which a concept, conclusion, or measurement accurately
reflects the real-world situation it is intended to represent.
• Validity establishes how well a technique, instrument or process measures a particular concept.
Factor analysis is usually used to establish the validity of a measuring instrument.
• Understanding validity is essential because it helps researchers and practitioners ensure that
their measurements are not only reliable but also meaningful in representing what they intend
to assess.
• High validity increases confidence in decision-making processes based on these measurements,
whether in testing, assessments, or contexts
 THE USE OF SOFTWARE PACKAGES FOR QUANTITATIVE
DATA ANALYSIS

• The use of statistics by researchers has been greatly facilitated by software packages
such as SPSS for Windows.
• A low-cost student version of SPSS (Statistical Package for Social Sciences), SPSS for
Windows, designed to run with Windows can produce a variety of tables as well as
rapidly undertake a variety of statistical analyses.

SPSS PRACTICALS
DISCUSSION
 ASSIGNMENT

Assignment III– Quantitative Research Methods for Social Sciences [100 marks]
• Answer must be typed in Times New Roman, 12 points, 1.5 line-spaced and the pages must be
numbered.
• Cover page must have full names and registration numbers for all group members, programme and
course name, level and research topic.
• Present the term paper:Assign III – Friday 4 October 2024
• Submit the presentation slides and word document – Thursday 3 October 2024.
Note that:
i. The term paper should employ a quantitative research design (quantitative research methods).
ii. Each group should choose a topic related to Conflicts Prevention and Early Warning
Mechanisms field.
Thank you!