06/02/2023

Research Methods
Data Processing & Analysis

 Learning Objectives
—) Identify elements of data processing.
—) Outline the data analysis methods & tools.
—) Select appropriate research software.
—) Understand the features of qualitative &
quantitative research.

1
 06/02/2023

 Data Processing and Presentation

– Processing implies editing, coding, classification and
tabulation of collected data so that they are amenable to
analysis.

 Editing
– Editing of data is a process of examining the collected raw
data (specially in surveys) to detect errors & omissions and
to correct these when possible.

– Editing is done to assure that the data are accurate,

consistent with other facts gathered, uniformly entered, as
complete as possible and have been well arranged to
facilitate coding and tabulation.

 Editing
– Editing should be done as soon as possible
• Preferable on the very day or on next day if it is an
interview.
• Use recording device

– Central editing should take place when all forms or

schedules have been completed and returned to the
office.
• Social science
• Regional data

2
 06/02/2023

 Coding
– Coding refers to the process of assigning numerals or
other symbols to answers so that responses can be put
into a limited number of categories or classes.

– Such classes should be appropriate to the research

problem under consideration.

– They must also possess the characteristic of

exhaustiveness and also that of mutual exclusively
(which means that a specific answer can be placed in
one and only one cell in a given category set).

 Coding
– Another rule to be observed is that of uni-
dimensionality by which is meant that every class is
defined in terms of only one concept.

– Coding decisions should usually be taken at the

designing stage of a data collection.

3
 06/02/2023

 Classification
– Classification is the process of arranging data in groups or
classes on the basis of common characteristics.

– Classification can be one of the following two types,

depending upon the nature of the phenomenon involved:
—) Classification According to Attributes
—) Classification According to Class-intervals

– Classification According to Attributes: data are classified on

the basis of common characteristics which can be
descriptive (such as literacy/educational level, sex……).

 Classification
– Classification According to Class-intervals: Unlike
descriptive characteristics, the numerical
characteristics refer to quantitative phenomenon
which can be measured through some statistical units.

– Numerical data relating to income, production, age,

weight….come under this category.

– Such data are known as statistics of variables and are

classified on the basis of class intervals.

4
 06/02/2023

 Tabulation
– When a mass of data ‘Big Data’ has been assembled, it
becomes necessary for the researcher to arrange it in
some kind of concise & logical order. This procedure is
referred to as tabulation.

– Thus, tabulation is the process of summarizing raw

data and displaying the same in compact form (i.e.,
in the form of statistical tables) for further analysis.

– In a broader sense, tabulation is an orderly

arrangement of data in columns and rows

 Tabulation
– Tabulation can be done by hand or by mechanical or
electronic devices.

– The choice depends on the size & type of study, cost

considerations, time pressures and availability of
tabulating machines or computers.

• Examples:
–MS-Excel Classification (sorting, filtering, grouping,
coding…..)
–Writing a Code/Machine Learning/Artificial Intelligence

5
 06/02/2023

 Data Processing and Presentation

- AKA Data Cleanup; Data Organization in Data
Mining/Data Science

- Example:
- Cutoff value
- >50% of missing value…..
- Imputation technique
- Replacement value – median, mean….
- Assign 2 Variable Responses’
- Yes & No - > Code it with 0 & 1

 Understanding the Data

– What is the type of data collected?
1. Categorical/Nominal
Example: Gender, marital status, nationality, color (hair/eye)….
2. Binary/Dichotomous
Example: Response to Yes or No type of questions….
3. Ordinal – Numbers in order or scale – sequence (can not do
arithmetic)
Example: Satisfaction, education level, economic status, letter
grade….
4. Continuous - fractions
Example: Age, speed, income, weight, height, time, score/grade…..
5. Discrete – can not be broken into decimal or fraction
Example: No. of children in a family, no. of players, days in a
week…..

6
 06/02/2023

 Understanding the Data

– What is the type of data collected?

 Data Analysis
– Is the process of inspecting, transforming, and modeling
data with the goal of discovering useful information
suggesting conclusions, & supporting decision making.

4 main types of data analysis

– Descriptive Analysis
• What happened? Overview
– Diagnostic Analysis
• Why did it happen? Road failure
– Predictive Analysis
• What is likely to happen in the future? Credit
– Prescriptive Analysis
• What is the best course of action? Roadmap

7
 06/02/2023

 Types of Data Analysis (Statistical Analysis)

– Statistical Data analysis can be made using:
i. Descriptive Statistics
ii. Inferential Statistics
– Descriptive statistics are used to describe, summarize,
or explain a given set of data.
– Inferential statistics is used to infer certain
characteristics of samples to population.

 Organizing and Presentation of Data

– Descriptive Statistics
– The followings are tools for organizing and presenting
data:
—) Frequency table
—) Histogram
—) Frequency polygon
—) Bar chart
—) Pie chart

8
 06/02/2023

 Descriptive Statistics
Measures of Central Measures of Dispersion
Other Methods
Tendency

Arithmetic Mean Range Ratios

Geometric Mean Mean Deviation Proportions

Median Variance Percentages

Mode Standard Deviation Rates of Change

Percentiles Z scores

Quartiles Coefficient of Variation

 Measures of Central Tendency

– Analysis carried out with a description of single variable
in terms of the applicable unit of analysis.

– The three most frequently used measures of central

tendency are:
—) Mode
—) Median
—) Mean

9
 06/02/2023

 Mode
– Mode can be defined as the most frequently occurring
value in a group of observations.

– If the scores for a given sample distributions are:

32, 32, 35, 36, 37, 38, 38, 39, 39, 39, 40, 40, 42, 45

– Then, the mode would be 39 because a score of 39

occurs three times, more than any other score.

– Mode is very good measure for ascertaining the

location of distribution in the case of nominal data.

 Median
– Median is defined as the middle value in an ordered
arrangement of observations.

– The median is often used to summarize the location of a

distribution.

– Further, the median can be used with ordinal, interval, or

ratio measurements.

– If the scores for a given sample distributions are:

32, 32, 35, 36, 37, 38, 38, 39, 39, 39, 40, 40, 42, 45
The median will be: 38 + 39 = 38.5
2

10
 06/02/2023

 Mean
– The arithmetic mean is the most commonly used &
accepted measure of central tendency.

– This should be used in the case of interval or ratio

data.

– If the scores for a given sample distributions are:

32, 32, 35, 36, 37, 38, 38, 39, 39, 39, 40, 40, 42, 45

– The mean of the distribution will be:

[32+32+35+36+37+38+38+39+39+39+40+40+42+45]/14 =
38

 Measure of Dispersion
– The measure of dispersion is as important as the
measure of location for data description and whenever
researchers describe the measure of location, they
should also specify the spread of distribution, which is
measured by measure of dispersion.

– The three most frequently used measures statistics

measuring variability and dispersion are:
—) Range
—) Variance
—) Standard deviation

11
 06/02/2023

 Range
– Range is the difference between the highest & lowest
value.
– It is based solely on extreme values. Thus, it cannot truly
reveal the body of measurement.
 Variance
– The squared deviation of a random variable from its mean.

– Variance makes deviation much larger than it actually is,

hence to remove the effect they are un-squared.
– Take the square root of the squared deviations in the process
of computing standard deviation.

 Standard Deviation
– The standard deviation provides the best measure of
dispersion for interval/ratio measurements and is the
most widely used statistical measure after mean.

– The standard deviation for a sample will be calculated

by the following formula:

– Formula to compute the standard deviation for

population:

12
 06/02/2023

 Example:
– The owner of a cafe is interested in how much people
spend at her cafe.

– She examines 10 randomly selected customers and

noted the following:
44, 50, 38, 96, 42, 47, 40, 39, 46, 50

– She calculated the mean by adding the observations and

dividing by 10 to get:
x = 49.2

 Example:
– Below is the table for getting the standard deviation:
X (X- 49.2) (X-49.2)2
44 -5.2 27.04
50 0.8 0.64
38 11.2 125.44
96 46.8 2190.24
42 -7.2 51.84
47 -2.2 4.84
50 -9.2 84.64
39 -10.2 104.04
46 -3.2 10.24
50 0.8 0.64
Total 2,600.4

13
 06/02/2023

 Example:
– Hence, the variance is 289 & the standard deviation is
the square root of 289 = 17.
– The mean for this example was about 49.2 and the
standard deviation was 17.
– We have:
49.2 - 17 = 32.2
49.2 + 17 = 66.2
– What this means is that most of the patrons probably
spend between 32.20 and 66.20.

 Empirical Rule
– For a Normal (Symmetric) distribution approximately:
—) 68% of the measurements fall within one standard
deviation (1SD) around the mean.

—) 95% of the measurements fall within two standard

deviations (2SD) around the mean.

—) 99.7% of the measurements fall within three

standard deviations (3SD) around the mean.

14
 06/02/2023

 Standard Deviation

 Skewness

Mean Mean Mean

Mode Mode
Median
Median Mode Median

Negatively Symmetric Positively

Skewed (Not Skewed) Skewed
Normal

15
 06/02/2023

 Research Software
– Statistical software are specialized computer programs
for statistical & econometric analysis.

– The most commonly used statistical packages in

research: Programming Languages
—) SPSS —) R
—) STATA —) MALTAB
—) SAS —) Python……
—) Minitab
—) NVivo

 What is SPSS?
– Originally it is an acronym of Statistical Package for the
Social Science, now it stands for Statistical Product and
Service Solutions.

– Developed by SPSS Inc. and later acquired by IBM

– SPSS 1 (1968) – SPSS 29 (2022)
– One of the most popular statistical packages which can
perform highly complex data manipulation and analysis with
simple instructions.
• Data management, advanced analytics, multivariate analysis,
business intelligence….

16
 06/02/2023

 Selecting Research Software

– Factors for selecting a tool for data analysis:
—) Is it user friendly?
—) Can you afford it?
—) Is it qualitative or quantitative type of data you want to
analyze?
—) Does it handle large enough data sets?
—) Does the software provide all the methods you need?
—) Does it run natively on your computer?
—) Does it fully support the style (programming, or menus and
dialog boxes, or workflow diagrams) that you like?
—) Does it provide output in the form you prefer?
—) Do your colleagues use it so you can easily share
data/programs?

 Qualitative Research
– Qualitative research is a research technique that allow a
researcher to obtain elaborate interpretations of phenomena
without depending on numerical measurements.

– The followings are characteristics of qualitative research:

—) Uses small than large samples
—) Emphasizes on unstructured method than structured
methods
—) Involves subjective interpretation rather than “objective”
statistical inference
—) Has an exploratory purpose rather than
descriptive/conclusive
—) Lack of data

17
 06/02/2023

 Qualitative Research
– Qualitative technique in business is usually undertaken in
the form of case study.
– Case study is a documented history of a particular person,
group, organization, or event.
– It is an intense examination of one or a few situations
typically:
—) Involves in-depth investigation and careful study
—) Requires cooperation from the investigated subjects
(cases)
– Case analyses are used to develop themes that can help
explain a phenomenon.

 Quantitative Research
– Relate to numbers
– Data must be able to be enumerated & presented in
terms of statistical patterns/associations

– Example:
• Cost growth calculations were determined as
positive or negative percentages using Equation 1
(Migliaccio et al. 2010):
𝐶𝑜𝑠𝑡 𝐺𝑟𝑜𝑤𝑡ℎ
𝐹𝑖𝑛𝑎𝑙 𝐶𝑜𝑛𝑠𝑡𝑟𝑢𝑐𝑡𝑖𝑜𝑛 𝑐𝑜𝑠𝑡 − 𝑂𝑟𝑖𝑔𝑖𝑛𝑎𝑙 𝐶𝑜𝑛𝑡𝑟𝑎𝑐𝑡 𝐶𝑜𝑠𝑡
=
𝑂𝑟𝑖𝑔𝑖𝑛𝑎𝑙 𝐶𝑜𝑛𝑡𝑟𝑎𝑐𝑡 𝐶𝑜𝑠𝑡

18
 06/02/2023

 Summary
– Data Processing is the process of editing, coding,
classification and tabulation of collected data so that they
are amenable to analysis.
– Descriptive analysis includes calculating measures of central
tendency (Mean, Mode and Median) and Measures of
variation (Rang, Variance and Standard Deviation).
– Qualitative research is a research technique that allow a
researcher to obtain elaborate interpretations of phenomena
without depending on numerical measurements as in
Quantitative research.
– There are many statistical software that help in data
analysis. Common software includes SPSS and STATA.

Research Methods
Hypothesis Testing

19
 06/02/2023

 Learning Objectives
—) Describe the features of hypothesis.
—) Test research hypothesis using parametric test.
—) Test research hypothesis using non-parametric test.
—) Use correlation and regression in data analysis.

 Hypothesis Testing
– Basic concepts in hypothesis testing
—) Generally in hypothesis testing, it is based on null
hypothesis.
—) Significance level is the maximum value of the probability
of rejecting the null hypothesis when it is true & usually
determined in advance before testing hypothesis.
—) A significance level of 5% or less is usually taken in
research.
—) Given a null hypothesis H0 and an alternative hypothesis
Ha, we make a rule which is known as decision rule
accept H0 (i.e., Reject Ha) or reject H0 (i.e., Accept
Ha).4

20
 06/02/2023

 Hypothesis Testing
– Two tailed test is appropriate when we have H0 : µ =
µH0 and Ha : µ ≠ µH0, which may mean µ > µH0 or µ <
µH0 . Thus, in a two-tailed test, there are two rejection
regions, one on each tail of the curve

– A one-tailed test would be used when we are to test,

say, whether the population mean is either lower than
or higher than some hypothesized value.

– Degree of freedom is the number of possibilities in your

choice minus one.

 Procedure for Hypothesis Testing

– Clearly state your hypothesis
– Selecting a significance level
– Deciding the distribution to use
– Computing an appropriate value
– Determine the degree of freedom
– Identify the table value
– Comparing the calculated value with table value
– Reaching on conclusion

21
 06/02/2023

 Parametric Hypothesis Testing

– Parametric tests usually assume certain properties of
the parent population from which we draw samples.
—) Observations should come from a normal population
—) The Sample size should be large enough
—) Population parameters like mean and variance must
hold good

– Important parametric tests are:

—) t-test
—) F-test
—) 2-test

 t-Test
– t-test is based on t-distribution and is considered an
appropriate for:
—) Judging the significance of a sample mean or for judging
the significance of difference between the means of two
samples in case of small sample(s) when population variance
is not known.
—) In case two samples are related, we use paired t-test for
judging the significance of the mean of difference between
the two related samples.
—) It can also be used for judging the significance of the
coefficients of simple and partial correlations.

22
 06/02/2023

 Independent Samples t-test

– Samples happen to be small samples and population
variances not known.

X1  X 2
t
 2 2
  
 (n1  1) s1  (n2  1) s2 n1  n2 
 n1  n2  2  
  n1n2  

– With degree of freedom (d.f.) = (n1 + n2 – 2)

 Illustration:
– Samples of daily sales of similar coffee shops in Bedele and
Agaro towns are taken for a new product with the following
results.
Towns Mean Variance Sample size
Bedele 57 5.3 5

Agaro 61 4.8 7
– Is there any evidence of difference in sales in the two towns?
– Use 5% level of significance for testing this difference
between the means of two samples.
– Table value of t (10,5%) = 2.228

23
 06/02/2023

 Solution:
– Hypothesis
Ho : 𝑋 = 𝑋 , Ha : 𝑋 ≠𝑋

X1  X2 57 61
t   3.053
 2   4x5.3  6x4.8 
 (n1 1)s1  (n2 1)s2  n1  n2  
2
 12 
 n1  n2  2 35 
  n1n2   

10


– Decision Rule: Reject Ho if ItI > 2.228

– Ho is rejected and Ha is Accepted
– There is a difference in average sales between Bedele
and Agaro

 Analysis of Variance (ANOVA)

– The significance of the difference between the means
of two samples can be judged through t-test, but
difficulty arises when we want to examine the
significance of the difference amongst more than two
sample means at the same time.

– ANOVA is essentially a procedure for testing the

difference among more than two different groups of
data.

24
 06/02/2023

 Types of ANOVA
– There are many types of ANOVA:
—) One way ANOVA
—) Two way ANOVA
—) Repeated measure ANOVA
—) Analysis of Covariance (ANCOVA)
—) Multivariate Analysis of Variance (MANOVA)

 One-Way ANOVA
– One-way analysis of variance is used to test the claim that
three or more population means are equal.
– This is an extension of the two independent samples t-test.
– The response variable is the variable you’re comparing.
– The factor variable is the categorical variable being used to
define the groups.
– The one-way is because each value is classified in exactly
one way
—) Examples include comparisons by gender, race, political
party, color and so on.

25
 06/02/2023

– In one way ANOVA

—) The null hypothesis is that the means are all equal.
—) The alternative hypothesis is that at least one of the
means is different.
– ANOVA doesn’t test that one mean is less than another,
only whether they’re all equal or at least one is
different.
Source SS df MS F p
Between
Within
Total

 Illustration
– Table below presents units of production per day by
three machines in four days. Test if the average unit of
production per day by the three machines are
statistically different at significant level of 5%.
Per day production by three machines in unit
Days
Machine A Machine B Machine C
1 6 5 5
2 7 5 4
3 3 3 3
4 8 7 4
Total 24 20 16

26
 06/02/2023

 Solution
Hypothesis Source SS df MS F P(2,9)
Ho: μA =μB= μC, H1: μA ≠μB Between 8 2 4 1.5 4.26

≠ μC Within 24 9 2.67
Total 32 11

𝑋 = (6+7+3+8)/4 =6
𝑋 = (5+5+3+7)/4 =5 Decision Rule: Reject Ho if
𝑋 = (5+4+3+4)/4 =4 F>P
𝑋 = (6+5+4)/4 =5
SSW = (6-6)2+(7-6)2+(3- Conclusion: There is no
significant difference
6)2+(8-6)2
among the mean of the
+(5-5)2+(5-5)2+(3- three machines.
5)2+(7-5)2
+(5-4)2+(4-4)2+(3-
4)2+(4-4)2

 Non Parametric Test

– Non parametric tests do not make any assumption about
the parameters of the population.

– Advantages of non parametric test:

—) Used With all Scales
—) Easier to compute
—) Make fewer assumptions
—) Need not involve population parameters
—) They may give as exact result as parametric models

27
 06/02/2023

 Non Parametric Test

– Disadvantages of non parametric test:
—) Non parametric models are not as efficient as
parametric models
—) Difficult to compute by hand for large samples
—) For non parametric models, tables are not widely
available

 Non Parametric Tests

1. Test of hypothesis to determine if categorical data
shows dependency or if two classifications are
independent (Chi-square test)

2. Test of a hypothesis concerning some single value for

the given data (such as one-sample sign test).

3. Test of a hypothesis concerning no difference among

two sets of data (such as U-test, W–test, two-sample sign
test, Fisher-Irwin test, Rank sum test and so on).

28
 06/02/2023

 Non Parametric tests

4. Test of a hypothesis of a relationship between variables
(such as Rank correlation, Kendall’s coefficient of
concordance and other tests for dependence).

5. Test of a hypothesis concerning variation in the given

data i.e., test analogous to ANOVA (Kruskal-Wallis test).

6. Tests of randomness of a sample based on the theory of

runs (one sample runs test).

 Parametric Versus Non Parametric Testing

Test Parametric Test Non Parametric Test
One Quantitative Response
One Sample t-test Sign Test
Variable
One Quantitative Response
Variable – Two Values from Paired Sample t-test Wilcoxon Signed Rank Test
Paired Samples
One Quantitative Response
Variable – One Qualitative Two Independent Wilcoxon Rank Sum or Mann
Independent Variable with two Sample t-test Whitney Test
groups

One Quantitative Response

Variable – One Qualitative
ANOVA Kruskall Wallis test
Independent Variable with
three or more groups

29
 06/02/2023

 Chi Square Test (2)

– 2 compares an observed frequency distribution with an
expected frequency distribution.

– Another common use of 2 is to determine whether

there is an association between two independent
variables.

– The formula to calculate 2 is given below:

 Conditions for the Application of (2) Test

– Observations recorded and used should be collected on
a random basis.

– 2 test is used after summarizing frequency distribution

in a table.

– All the items in a sample must be independent.

– No group should contain very few items, say less than
10.

– The overall number of items must also be at least 50,

howsoever small the number of groups may be.

30
 06/02/2023

 Illustration
– Table below shows the data obtained during accounting
research from 250 firms. Test the association between
tax payers category and the choice of recording systems
using 2 at 5% level of significance and determine the
degree of association.

– Table Value F(2,5%)=

Single 5.991
Entry Double Entry Total
Category A 10 40 50
Category B 30 50 80
Category C 60 60 120
Total 100 150 250

 Solution
– Ho: There is no association between tax payers category
and the choice of recording systems.
– Ha: There is an association between tax payers category
and the choice of recording systems.
O E O-E (O-E)2 (O-E)2/E
10 20 -10 100 5.0
40 30 10 100 3.3
30 32 -2 4 0.1
Decision Rule: Reject Ho if 2 > 5.991
50 48 2 4 0.1
60 48 12 144 3.0
Conclusion: 60 72 -12 144 2.0
Reject Ho and Accept Ha 2 13.5
There is an association between tax
payers category and the choice of
recording systems.

31
 06/02/2023

 Correlation and Regression Analysis

– Correlation analysis is used to describe the strength &
direction of the linear relationship between two
variables.
– The followings are types of correlation analysis from
which researchers can chose.
—) Pearson product Moment correlation
—) Spearman Rank order correlation
—) Point biserial correlation
—) Phi correlation
—) Multiple correlation
—) Partial correlation

 Regression Analysis
– Regression analysis is used for prediction.
– It can be simple or multiple regression.
– The different types of multiple regression are given
below. Multiple
Regression
Models
Non-
Linear
Linear

Dummy Inter-
Linear action
Variable

Poly- Square
Log Reciprocal Exponential
Nomial Root

32
 06/02/2023

 Multivariate Analysis
– Multivariate statistics are increasingly popular
techniques used for analyzing complicated data sets.
– They provide analysis when there are many
independent or many dependent variables that are
correlated with one another to varying degrees.

 Multivariate Analysis
– Multivariate statistics are an extension of uni-variate
and bivariate statistics.
– The trick in multivariate statistics is not in
computation; that is easily done by computer.
– The trick is to select
• Reliable & valid measurements
• Choose the appropriate program
• Use it correctly
• Know how to interpret the output

33
 06/02/2023

 Types of Multivariate Analysis

– Multiple regression
– Canonical correlation
– Discriminant analysis
– Factor analysis
– Path analysis
– Analysis of covariance structures
– Multivariate analysis of variance
– Cluster analysis
– Multidimensional scaling
– Latent structure analysis

 Other Analysis
– Artificial Neural Network (ANN)
– Fuzzy Logic
– Structural Equation Modeling (SEM)
– Time Series Analysis
– Optimization
• Queuing Theory, Linear & non-linear Programing, Ant Colony,
Genetic…..
– Root Cause
• Fault Tree Analysis (FTA), Failure Mode Effect Analysis
(FMEA)….
– Artificial Intelligence (AI)
–

34
 06/02/2023

 Summary
– Researchers usually formulate two types of hypothesis
in their study. This includes null hypothesis and
alternative hypothesis

– Parametric tests usually assume certain properties of

the parent population from which we draw samples.

– The important parametric test examples: t-test and F-

test.

 Summary
– Non parametric tests do not make any assumption about
the parameters of the population such as U-test, W–
test, two-sample sign test, Fisher-Irwin test, Rank sum
test and so on.

– Correlation analysis is used to describe the strength and

direction of the relationship between variables.

– Regression analysis is used for prediction. It can be

simple or multiple regression.

35
 06/02/2023

Research Methods
Interpretation

 Learning Objectives
—) Understand the methods of interpreting result.
—) Outline elements of research report.
—) Communicate research result through different
means.
—) Understand how to be ethical in research
undertaking.

36
 06/02/2023

 Data Interpretation
– Data interpretation is adding meaning to the result by
making connections and comparisons and exploring
causes and consequences.
– Things to consider when interpreting your data:
—) Interpret findings based on the purpose/objectives
of your study.
—) Relate the findings to real life context.
—) Use persuasive language to convince your readers to
see the research from your point of view.
—) Order your interpretation to highlight important
findings.

 Positive and Negative Results

– When results are positive and the data support the
hypotheses, one interprets the data along the lines of
the theory and the reasoning behind the hypotheses.

– If we can repeat the achievement, then the evidence of

adequacy is even more convincing.

– If we can be fairly sure that the methodology, the

measurement, and the analysis are adequate, then
negative results can have definite contributions to
scientific advancement.

37
 06/02/2023

 Unanticipated Findings
– Unpredicted relation may be an important key to a
deeper understanding of the theory.

– Unpredicted and unexpected findings must be treated

with more caution than predicted and expected
findings.

– Before being accepted, they should be substantiated in

independent research in which they are specially
predicted and tested.

 Proof, Probability and Interpretation

– A researcher’s aim is to bring evidence to bear that
such-and such a proposition is true.

– Proof is a deductive matter. Experimental methods of

inquiry are not methods of proof, they are controlled
methods of bringing evidence to bear on the probable
truth or falsity of relational propositions.

– In short, no single scientific investigation ever proves

anything. Thus, the interpretation of the analysis of
research data should limit the use of the word proof.

38