Research Methodology for your project

Hypothesis and analysis of data

Research Process
1. 2. 3. 4. 5. Research Problem Review of Literature Hypothesis development Sampling design Data Collection
i. Method ii. Measurement

6. Analysis and interpretation 7. Research Report

Scales of Measurement

Variable
A variable can be defined as anything that can or does change. Change Within, between Types Nominal Ordinal Interval and ratio variables
4

Scales of Measurement
Various scales are used in measurement 4 Types
1. 2. 3. 4. Nominal Scale or categorical variable Ordinal Scale or Rank variable Interval Scale or Scaled variable Ratio Scale

Why scaling important

The kinds of descriptive statistics and significance tests that are appropriate depend on the level of measurement of the variables concerned ie., nominal, ordinal, interval, or ratio.

Nominal Scale
Numbers used to categorise objects or events. Numbers are interchangeable
(1) Eg : Gender : 1 for male & 2 for female (2) Eg. : Numbering Athletes Jerseys The players on a Football team have numbers on their jerseys: e.g. 2, 6,10, 12, etc.

E.g. religion, area of residence, occupation, store types, dept, etc The rule is: Do not assign the same numeral to different classes or different numerals to the same class . Beyond that, anything goes with the nominal scale.

Statistical Tools
Descriptive Statistical techniques :
Mode, Frequency tables & percentage analysis

Inferential : Cross tabulation and chi square

Ordinal Scale
Numbers used to rank items. Determination of greater or less.
Hierarchical order

Ranks are not interchangeable, unequal intervals

Egs: Ranking of brands, Other instances are found among scales of intelligence, measuring education by degrees attained, i.e. bachelors, masters, or doctorate.

Statistics : - Median, Mode, Percentages, frequency tables Rank correlation, Cross tabulation NB : Arithmetic mean (or average) should not be used on the ordinal scale.
8

Nominal and Ordinal Scales

Nominal and ordinal scales are considered to be lower level of scales.
(higher levels of scales interval and ratio)

The branch of statistics that deals with nominal and ordinal measurement is called non parametric statistics.

Interval Scale
Quantitative scale classification, order, & equal intervals
Most psychological tests, measures of attitude, personality, and the like, are interval measures. Business Research : Satisfaction level, preference rating, Attitudes

Analysis : To analyse response in Likert scale, each response is assigned a numerical value.
Item by item basis Single score for each respondent

NB : High (Low) score must consistently reflect a favourable response. Statistical Technique : All advanced tests
t test, ANOVA, regression, Factor analysis,.....
10

Ratio Scale
Scales on which the value of zero means none , or the total absence of the variable, are ratio scale of measurement. Eg:- Measures of physical dimensions weight, height,
distance Business research :- Sales, profit, ROTA, income.
11

Properties of Four scales

Scale Nominal Ordinal Classific ation YES YES Order NO YES Distance NO NO Origin NO NO Examples Gender, Store type Ranking for brands, Market position Attitudes, opinions, Many psychological measures Age in years, income in Rs, Sales, Costs , No of consumers, Market share

Interval

YES

YES

YES

NO

Ratio

YES

YES

YES

YES

Source: Adapted from U. Sekaran.(2006). Research Methods for Business. (4th ed.). New Delhi : Wiley India (P) Ltd, p.189.

Scales and Statistical Techniques

All statistical techniques applicable to a given scale are also applicable to any higher scale.
For eg. All statistics applicable to an ordinal scale are also applicable to interval and ratio scales.

13

Reducing scales of Measurement

If a variable can be measured on an interval or ratio scale, it can also be reduced and measured on an ordinal scale. To simplify things by reducing interval or ratio data to ordinal categories. Egs

14

Scales and Descriptive Statistics

Scale Nominal Ordinal Interval Measure of central Tendency Mode, percentages Median Mean, Range Arithmetic mean, Geo metric Mean, Harmonic Mean
-

Measure of Dispersion

Percentiles Standard Deviation Standard Deviation or Variance or coefficient of variation

Ratio

Scales and statistical test

Level of Data No of Samples Independent Samples? Sample Size Appropriate Statistical Technique

Interval Interval Interval Interval Interval Nominal

2 2 2 2+ 2+ 2

Yes Yes No Yes No Yes

30 <30 a a a a

Z test t test Paired sample t test One way ANOVA

Paired sample t test of all pairs

chi square test

a : Sample size is not a determinant of the appropriate statistical technique

Source :Adapted from D. S. Tull.,& D.I. Hawkins.(1993). Marketing Research.(6th ed.). New Delhi: Prentice Hall of India Ltd. p.629.

Small sample and large sample

If sample size is <30 :- Small sample If sample size is 30 :- Large sample

17

Independent samples & Related samples

Indept samples (Equal or unequal sample size)
Eg. Male and female samples

Related samples (Equal sample size)

Eg. (i) Husband and wife (ii) Father, mother and child (iii) Before and after data same sample

18

Data Analysis
Tabulation of data (Excel) Reliability of the instrument (if interval scaled)
Crobanch s alpha - SPSS

Descriptive statistics and charts (Excel) Hypothesis Testing (SPSS)

Crobanch s alpha
Used to describe the reliability Alpha coefficient ranges in value from 0 to 1 The higher the score, the more reliable the generated scale is. 0.7 is considered to be an acceptable reliability coefficient but lower thresholds are sometimes used in the literature.

Hypothesis
A hypothesis is a precise testable statement A prediction of what the researcher expects to find or prove Types
Alternate hypothesis
is the expected conclusion proposition that is accepted if the null hypothesis is rejected.

Null hypothesis
is the hypothesis being tested. No difference/no relationship

Hypothesis Development in your research

Two variables
One demography and one from your research question Two research questions

Relation or Difference between two variables

Hypothesis e.g???
Difference hypotheses Relation hypotheses

Hypothesis testing
Develop alternate and null hypothesis Hypothesis tests

Hypothesis Tests
Difference
One nominal variable and an interval scale
t test (if there are only 2 groups)
1. 2. Independent Samples t test Paired Sample t test

Anova (if there are more than two groups)

Relationship/ Association
Between two interval scaled variables
Regression analysis

Between two nominal/ordinal variables

Chi square

Hypothesis Tests in SPSS

1. Import data from EXCEL to SPSS 2. Give values for nominal variables 3. Hypothesis testing

About SPSS
SPSS is a computer program used for statistical analysis.
It is used by market researchers, health researchers, survey companies, government, education researchers, marketing organizations and others.

SPSS (originally, Statistical Package for the Social Sciences) was released in its first version in 1968. The company SPSS Inc. announced on July 28, 2009, that it was being acquired by IBM for US$1.2 billion. As of January 2010, it became "SPSS: An IBM Company".
No abbreviation now

SPSS Versions
Version 10 (SPSS-X) in 1983 SPSS 15.0.1 - November 2006 SPSS 16.0.2 - April 2008 SPSS Statistics 17.0.1 - December 2008 PASW Statistics 17.0.3 - September 2009 PASW Statistics 18.0 - August 2009 PASW Statistics 18.0.1 - December 2009 PASW Statistics 18.0.2 - April 2010 PASW Statistics 18.0.3 - September 2010 IBM SPSS Statistics 19.0 - August 2010

Statistics and SPSS

Statistics included in the base software: Descriptive statistics:
Cross tabulation, Frequencies, Descriptives, Explore, Descriptive Ratio Statistics

Bivariate statistics:
Means, t-test, ANOVA, Correlation, Nonparametric tests

Prediction for numerical outcomes:

Linear regression

Prediction for identifying groups:

Factor analysis, cluster analysis (two-step, K-means, hierarchical), Discriminant analysis

Statistical Terms
Degrees of freedom, level of significance, reliability and validity

Degrees of freedom (df)

In statistics, the number of degrees of freedom is the number of values in the final calculation of a statistic that are free to vary. The number of independent pieces of information that go into the estimate of a parameter is called the degrees of freedom (df). In general, the degrees of freedom of an estimate is equal to the number of independent scores that go into the estimate minus the number of parameters estimated as intermediate steps in the estimation of the parameter itself (which, in sample variance, is one, since the sample mean is the only intermediate step).

Level of Significance
The probability that observed or greater differences occurred by chance. In statistics, a result is called statistically significant if it is unlikely to have occurred by chance - test of significance The amount of evidence required to accept that an event is unlikely to have arisen by chance is known as the significance level or critical p-value If the obtained p-value is small, then it can be said either the null hypothesis is false or an unusual event has occurred.

Reliability and Validity

How precise are your measurements? The two most important aspects of precision are reliability and validity. Reliability refers to the reproducibility of a measurement. Validity of a measurement tool (i.e. test in education) is considered to be the degree to which the tool measures what it claims to measure. Validity is often assessed along with reliability

t test and Levene s test

Levene's test is often used before a comparison of means. Levene's test is an inferential statistic used to assess the equality of variances in different samples.
Some common statistical procedures assume that variances of the populations from which different samples are drawn are equal. Levene's test assesses this assumption.

It tests the null hypothesis that the population variances are equal (called homogeneity of variance).
If the resulting p-value of Levene's test is less than some critical value (typically 0.05), the obtained differences in sample variances are unlikely to have occurred based on random sampling. Thus, the null hypothesis of equal variances is rejected and it is concluded that there is a difference between the variances in the population.