DATA ANALYSIS
QUANTITATIVE AND QUALITATIVE
(WITH THE HELP OF SOFTWARES)
Statistical Data Analysis
with software
(quantitative)
What is data analysis
• Data Analysis refers to the process of ordering,
structuring and treating the data to obtain
results. It addresses key features of the data and
their use and involve application of some
qualitative and quantitative techniques. It's a
part of the chain between data collection,
representation and interpretation. GIS and
computers are technological tools for data
display and analysis.
Statistical Data Analysis (montello)
• Data analysis is the set of display and mathematical
techniques and attendant logical and conceptual
considerations that allow us to extract meaning from
systematically collected measurement of our
phenomena of interest and communicate it to others.
• Data analysis helps in description, prediction,
explanation and control.
• Data analysis helps to efficiently identify and describe
pattern in large amount of data pattern, patterns that
improve our prediction about the unknown, explain the
causes of relationship and allow us to expand influence
over phenomena we wish to control.
Basis of statistical data analysis
• In statistical Approach to any data, including spatial
data, three basics are involved: (G.krishan, p 113)
• Classification- weather type, functional classification
of towns, frequency distribution
• Ordering- where research problem involve hierarchies
• Variable precision- exact measurement of any
phenomenon.
Type of analysis of statistical data
(Montello)
• Statistical Description
• Statistical Inference
• Estimation and Hypothesis testing
• Geospatial analysis
Statistical Description
• Statistical description involves summarizing and
organizing data to provide insights into its basic
characteristics, which are parameters. This type of
analysis uses measures of central tendency (mean,
median, mode), measures of dispersion (range,
variance, standard deviation), and graphical
representations (histograms, bar charts, scatter plots)
to present data in a coherent and understandable
manner. Finding relationships, for example correlation
and regression analysis.Descriptive statistics are
essential for providing an overview of datasets before
proceeding with more complex analyses, making them
foundational in research and decision-making.
• For the whole data
• Derived scores- percentile, z score
Statistical Inference
• Statistical inference involves drawing conclusions about
a population based on a sample of data. It uses
techniques such as confidence intervals, significance
tests, p-values, and regression analysis to make
predictions or generalizations. Methods like t-tests, chi-
square tests, ANOVA, and correlation analysis are
commonly employed to assess relationships between
variables or to test hypotheses. Statistical inference is
crucial when it is impractical or impossible to study an
entire population, allowing researchers to make reliable
conclusions from
Estimation and hypothesis testing
• Estimation and hypothesis testing are core components
of inferential statistics. Estimation focuses on
determining the approximate value of a population
parameter based on sample data, using point estimates
and interval estimates. Hypothesis testing, on the other
hand, involves formulating a null hypothesis and an
alternative hypothesis, then using statistical tests (e.g.,
z-test, t-test, chi-square test, ANOVA) to determine
whether there is sufficient evidence to reject the null
hypothesis. These techniques are essential for
validating research findings and making data-driven
decisions.
Geospatial Analysis
• Geospatial analysis involves collecting, manipulating, and
analyzing spatial or geographic data to identify patterns,
relationships, and trends across different locations. This type of
analysis integrates statistical methods with spatial data, enabling
researchers to explore how phenomena vary over space.
Common techniques include spatial statistics like Moran's I
(measuring spatial autocorrelation), Getis-Ord Gi* (detecting hot
spots and cold spots), Spatial Regression (examining
relationships between variables while accounting for spatial
dependence), and Kernel Density Estimation (creating smooth
density surfaces from point data). Additionally, methods like
Spatial Interpolation (e.g., Kriging, Inverse Distance Weighting)
are used to estimate values at unsampled locations.
• Geospatial analysis is often conducted using Geographic
Information Systems (GIS) software such as ArcGIS, QGIS, and
Google Earth Engine, which provide tools for mapping,
visualization, and spatial modeling. This type of analysis is
essential in various disciplines, including geography, urban
planning, environmental science, epidemiology, public health,
and transportation. For example, it is used to analyze disease
outbreaks, assess environmental risks, optimize urban
infrastructure, and model spatial patterns of social or economic
phenomena. By combining spatial data with statistical
techniques, geospatial analysis provides powerful insights that
are often invisible through traditional non-spatial statistical
methods.
Softwares (Cresswell)
• R : The R Project for statistical computing
• SPSS
• Microsoft Excel
• JPM
• Minitab
• Systat 13
• SAS/STAT
• Statarelease 17
Example of SPSS
Qualitative Data Analysis
using software
What is qualitative data analysis
Qualitative research techniques, such as
interviewing and participant observation,
are commonly used by human
geographers.
They generate rich data in the form of
interview transcripts and diaries.
Traditionally, geographers have analysed
this material by coding it manually by pen
and paper but this approach has a number
• The problems in the use and analysis of qualitative material
include :
• the volume of data generated;
• the complexity of data analysis;
• and the notion that clerical tasks required for data preparation and
management may prevent thorough data analysis.
• As a result in recent years, computer packages have
increasingly been adopted for qualitative data analysis
because they are regarded as useful tools for handling and
coding large amounts of written data and because they
facilitate the in-depth examination of relations between and
within these data
What is qualitative data analysis
• Steps in qualitative data processing-( Cresswell)- p207
• Organizing and preparing data for analysis.
• Going through all the data,
• coding the data.
• Identifying themes
• Developing a storyline interpretation.
• Further analyzing the data using analytical framework.
• Representing and interpreting data.
Coding- (Clifford, Cresswell)
• Coding is basically a way of evaluating and organizing data in an
effort to understand meanings in a text. First, coding helps the
researcher identify categories and patterns.
Types of Codes
• Strauss---four types of themes in the data: conditions, interaction
among the actors, strategies and tactics and consequences.
• Many of these are indicated to the researcher directly by the
subject. ‘Conditions’ can be indicated by such phrases as
‘because’ or ‘on account of or passages like ‘when I was in that
situation …’. For example, if a low-income mother referred to a
time when she lost her housing and had to live with her sister, we
would code it as a particular type of housing ‘condition’.
• Similarly, ‘interaction among the actors’ means looking for how
the informants engage with others, what they think of others, what
others do to them. So we might be interested in how low-income
mothers interact with their children, their partners, their
caseworkers, their employers and so on.
• ‘Strategies and tactics’ refers to what people do in certain
situations or how they handle particular events. For example, poor
women with children often exchange favours with friends and
neighbours as astra tegy to survive on low wages and/or state
benefits
• Finally, ‘consequences’ are often easy to identify because the
informant makes the connection for us: a woman says, ‘My child
was sick so I missed work too many times and lost my job.
Types of Codes
• Expected-( based on literature) and Sursprising ( while
conducting study)
• Descriptive/InVivo codes (in language of the
interviewee) and Analytical codes researcher’s insight
leading to theme building
• )(Clifford, p 447, the table 27.1)
Predetermined Code and Code book
(Cresswell)
• Code can be predetermined or emerge during data
analysis
• Code book give instruction of meaning of the code and
how and when to use a code
See patterns in coding
• Frequency
• Similarities
• Difference
Theme Building
• The coding process is fluid and dynamic but it is not the
end-product of analysis. As codes become more
complex and more connected to the project’s
theoretical framework, they start building into themes
that can then serve as the main topics for the final
product
• The process of theme building is central to qualitative,
interpretive work because it allows for the organization
of information into trends, categories and common
elements that are theoretically important. Themes may
be based on similarities within the data or, conversely,
on differences that appear and are interesting for some
reason.
• There’s the distinction between inductive and deductive
approaches:
• An inductive approach involves allowing the data to
determine your themes. (predetermined codes)
• A deductive approach involves coming to the data
with some preconceived themes you expect to find
reflected there, based on theory or existing
knowledge. (emerging codes)
Software assisted analysis-clifford
Three types of CAQDAS can be distinguished according to their key
functions:
(1) text retrievers/textbase managers-in brief, find words or
combinations of words in so-called string searches and put these
into new files.
(2) coding and text retrieval-They are used to mark text segments
for coding, cut and sort text segments, and find, retrieve and
report on data.
(3) theory builders-have the capacity to test hunches, ideas and
hypotheses that are pre-existing or emerge from data in the
project. They allow the researcher to make connections not only
between code and data but also between code and code.
Relationships can often be displayed in a hierarchical index
system or a graphic network.
Example of softwares
features Of NVivo20
