•Introduction to Geographical Research

Lecture 1

•Lecture Outline

•Basic concepts and definition

•Problem facing geographical data collection

•Scale of measurement

•Introduction

•This course is about techniques of collecting, processing, analysing and presenting geographical data.

•Also it deals with the techniques used in generalizing sample results into population and quantifying the associated risks.

•To start with let’s discuss the basic concepts in geographical research.

•Research

•Research is a process of enquiry and discovery.

•In geography, research is the process of trying to gain a better understanding of the relationship between humans, space, place and the environment and advance our understanding of our interactions with the world.

•Good research occurs at the intersection between theory, method, and practice.

•Cont…

•Research involves asking one or more questions and devising a plan for finding out answers (which may be provisional, varied, and even contradictory).

•Research should contribute to knowledge and understanding about something.

•Researchers seek variously to explore, describe, explain, and predict. ( Kitchin and Tate, 2000)

•Cont…

•It can also be defined as a scientific and systematic search for gaining information and knowledge on a specific topic or phenomena.

•On evaluating these definitions we can conclude that research refers to the systematic method consisting of:

•Cont…

1.Enunciating the problem

2.Formulating a hypothesis and/or research questions,

3.Collecting the fact or data,

4.Analysing the data and

5.Reaching certain conclusions either in the form of solutions towards the concerned problem or in certain generals for some theoretical formulation.

•Geographical Research

•The field of geographical research is not only a physical location or domain, but also includes the social terrain and the encounter of people with their environment.

•“Being there” is a vital dimension of geographic research; but at the same time raises questions of perspective – the “politics of ‘being there’”; issues of ‘naming and claiming’

•Cont…

•Geography researchers need to navigate access to spaces and cultures and acknowledge our own biases, preconceptions, prejudices, and limitations. There is no such thing as an omniscient, detached, objective researcher.

•Successful field work relies on the senses, especially developing our capacity to see and listen.

•Cont…

•Geographical knowledge is constructed and interpreted.

•The questions you ask influence the answers you get (Cloke et al, 2004)

•Aims of Geographical Research

1.Identifying general patterns or relations

2.Testing or refining existing theories

3.Making predictions (or prescriptions), or diagnosing something

4.Interpreting culturally or historically (or politically or economically or spatially) significant phenomena

5.Explaining social diversity

6.Giving voice to those who are outside the mainstream

7.Advancing new theories

(adapted from Hogart et al, 2002)

•Research Methods

•Research methods refers to techniques used in conducting research.

•Research methods are divided into three groups namely:

1.methods concerned with collection of data,

2.Statistical techniques used to establish relationships between variables and 

3.Method used to evaluate the accuracy of the result.

•Research Problem

•Research problem is perceived gap in knowledge that is helpful in problem solving.

•It is the real world phenomena or event for which existing theories or laws cannot describe, explain or predict.

•Knowledge/Research Gap

•Knowledge gap is missing or additional knowledge required to be able to describe, explain or predict a real world phenomena or an event.

•It describe the situation in which real world phenomenon or event cannot be described , explained or predicted in the light of existing theories.

•Cont…

•The easiness to fill the knowledge gap depends on the extent to which a problem situation is structured or unstructured.

•When the problem situation is structured, the components with their connections can easily be identified and modelled. 

•Situation in natural sciences are more structured and therefore more predictable than in social sciences.

•Scientific Theories

•Scientific theories are systematically organized set of facts and their logical relationships such that they are able to explain and predict new facts and events.

•It is based upon proven hypothesis and verified multiple times by group of researchers.

•Theory implies that something has been proven and is generally accepted being true.

•Cont…

•In other words a theory is in position to state that “if x then y”.

•Scientific theory are relative rather than absolute, because they change and are being modified to suit new facts, events and relationships.

•Scientific Hypothesis

•Scientific  hypothesis is a relative and logical answer to questions posed in a research.

•Once tested and confirmed by experiment hypothesis becomes a theory.

•It is an educated guess but which has not been proved.   

•Nature of Data

•Data is a raw information about various variables (attributes) that are collected in a research or study.

•Data becomes information when analyzed and become relevant to the problem under investigation.

•Information becomes fact when there is data to support.

•Classification of Data

•Data may be classified as:

1.Quantitative (numerical) or Qualitative data

2.Univariate, bivariate or Multivariate data

3.Primary or secondary data

•Univariate Data

•Univariate data set is the data set consisting observations on a single attribute or variable.

•They are divided into categorical and numerical data set.

•Categorical (qualitative) if the individual observations are categorical responses. For example sex of a person.

•Numerical (quantitative) data set if the observations by counting or measurements are number such as height of a person.

•Bivariate Data

•Bivariate data is the data consisting the observation on two attributes such as  age and height when measured simultaneously.

•Bivariate data set is also divided into categorical and numerical data.

•Multivariate Data

•Multivariate data set is the data set consisting two or more observation such as age, height and weight. Also multivariate data set is divided into categorical and numerical data.

•Primary and Secondary Data

•Secondary data are pieces of information that have already been collected for a different purpose, but may be relevant to the research problems at hand.

•Primary data, in contrast, are survey, observation, or experimental data collected to address the problem currently under investigation.

•Types of Numerical Data

Discrete Data

•Discrete data refers to data which possible values are isolated points on the number line. This type of data cannot be written in fractions. It is obtained by counting such as 1, 2, 3, etc.

Continuous Data

•Continuous data refers to data which possible values form an entire interval in the number line. This type of data can be written in fractions or decimal points. It is obtained by measurements such as 10.6 km, 0.7 kg.

•Types of Data Variable

•A variable is defined as characteristics of a person, object or phenomenon that takes two or more forms that is it has to vary for example height of different people.

•Variables can be classified quantitative or qualitative. Quantitative variables consist of variables which their result or observation can be measured e.g. height.

•Cont…

•Qualitative variable refers to the variables which numerical measurement is not possible such as colour and sex.

•There are several types of data variable as follows:-

•Background Variable

•Background variable describe socio-demographic characteristics of the respondents e.g. age, sex and marital status.

•They should be included in most social research. It help to describe relationship between respondents and other variables.

•Independent Variable

•Independent variable consist of predictor or causative variable.

•It is the factors which researcher thinks it explain variation in the dependent variable.

•Dependent Variable

•Dependent variable is the one which influenced by other variables especially the independent variable.

•Usually the researcher attempts to predict or explain the variations in the dependent variable.

•Intervening Variable

•Intervening variables are the ones which intervene the relationship between the independent and dependent variable.

•They are secondary independent variable. Example; The impact of fuel on transport cost.

•Independent variable: price of fuel, Dependent variable: transport costs, Intervening variable: fuel production.

•Extraneous Variable

•Extraneous variable refers to variable other than the independent variable that influence the dependent variable.

•Due to extraneous variable it is difficult to conclude that independent and dependent variable have a direct relationship.

•Example: In a study to determine the impact of English language to performance of students.

•Cont…

•If teachers using English language are more qualified than those using Kiswahili or other language.

•Then the performance of students could be due to qualification of their teachers and not due to the English language.

•Extraneous variable: Quality of teachers.

•Confounding Variables

•Confounding variable refers to extraneous variable that vary systematically with the independent variable.

•Example: If student who attend tuition are more intelligent than those who did not attend tuition then performance could be due to intelligence levels or combination of intelligence and tuition.

•Problems of Geographical Data Collection

•Collection of geographical data faced with number of problems.

•Theses problems depend on the source of data weather primary or secondary.

•To start with lets discuss problem facing collection of primary data

•Problems in Primary data collection

1.Language barrier

2.Change of the behaviour of the respondent

3.Unwillingness of the respondents to answer questions especially in interview or on questionnaire.

4.Inadequate resources

5.Illiteracy

6.Remoteness of the area

•Cont…

7. Problem of Sampling Frame

8. Weather changes

9. Wastage of time.

•Problem Facing Secondary Data Collection

1.Inaccessibility of data – some governmental documents may be confidential.

2.Poor quality of data

3.Data may be outdated

4.Data may be inadequate

5.Data may be aggregate

•Scale of Measurement

•The scale of measurement is a level by which a variable is measured. There are three things need to be considered when we study scale of measurement.

•First, any thing that can be measured falls into one of the four types. Second, the higher the scale of measurement, the more precision in measurement and third, every level up contains all the properties of the previous level.

•Cont…

•The four scales of measurement, from lowest to highest, are as follows:

1.Nominal

2.Ordinal

3.Interval

4.Ratio

•Ordinal and nominal data are always discrete. Continuous data has to be at either ratio or interval scale of measurement.

•Nominal Scale

•Nominal variables include demographic characteristics like sex, race and religion.

•The nominal scale of measurement describes variables that are categorical in nature.

•Ordinal Scale

•The ordinal scale of measurement describes variables that can be ordered or ranked in some order of importance.

•It describes most judgements about things, such as big or little, strong or weak.

•Most opinion and attitude scales or indices in the social sciences are ordinal in nature.  

•Interval Level

•The interval scale of measurement describes variable that have more or less equal intervals, or meaningful distances between their ranks.

•For example, if you were to ask somebody if they were first, second or third generation immigrant, the assumption is that the distance, or number of years, between each generation is the same.   

•Ratio Level

•The ratio scale of measurement describes variables that have equal intervals and fixed zero (or reference) point.

•It is possible to have zero income, zero education and no involvement in crime, but rarely do we see ratio level variables in social sciences since it’s almost impossible to have zero attitudes on things, although “not at all”, “often”, and “twice as often” might qualify as ratio scale of measurement. 

•Advanced statistics require:

1.At least interval scale of measurement, so the researcher always strives for this level,

2.Accepting ordinal level (which is the most common) only when they have to.

3.Variables should be conceptually and operationally defined with levels of measurement in mind since it’s going to affect the analysis of data later.

•Seminar Question

•With vivid examples examine the problems facing geographical data collection and suggest solutions for those problems.