Statistics

An Introduction and Overview In Research Process

Dejene Girma (PhD) Dilla University Department of Mathematics

 Statistics

We use statistics for many reasons:

 To mathematically describe/depict our findings
 To draw conclusions from our results
 To test hypotheses
 To test for relationships among variables

Numerical representations of our data

Can be:
 Descriptive statistics summarize data.
 Inferential statistics are tools that indicate how much confidence we can

have when we generalize from a sample to a population.

Dejene Girma (PhD) Dilla University Department of Mathematics
 Statistics

Powerful tools… we must use them for good.

 Be sure our data is valid and reliable
 Be sure we have the right type of data
 Be sure statistical tests are applied appropriately
 Be sure the results are interpreted correctly
 Remember… numbers may not lie, but people can

The proper care and Feeding of Statistics

Sampling & Statistics :
 Statistics depend on our sampling methods: Probability or Non-probability?
(i.e. Random or not?)
Dejene Girma (PhD) Dilla University Department of Mathematics
 Probability Samples

Even with probability samples, there is a possibility that the statistics we

obtain do not accurately reflect the population.
 Sampling Error : Inadequate sampling frame, low response rate,

coverage (some people in population not given a chance of selection)

 Non-Sampling Error : Problems with transcribing and coding data;

observer/ instrument error; misrepresentation as error.

Dejene Girma (PhD) Dilla University Department of Mathematics

 Measurement

 Levels of Measurement – the relationship among the

values that are assigned to a variable and the attributes of
that variable.
Levels of Measurement
 Nominal- naming
 Ordinal- rank order (high to low but no indication of
how much higher or lower one subject is to another)
 Interval- equal intervals between values
 Ratio- equal intervals AND an absolute zero (i.e. a
ruler)
Dejene Girma (PhD) Dilla University Department of Mathematics
 Levels of Measurement

Dejene Girma (PhD) Dilla University Department of Mathematics

 ….Levels of Measurement

EXAMPLE
 Age: under 30, 30-39, 40-49, 50-59 ---………………….--Interval
 Gender: Male, Female …………………………………………….
….Nominal
 Level of Agreement: Strongly Agree, Agree, Neutral,
Disagree, Strongly Disagree
………………………………………………………..…… Ordinal
 Percentage of the library budget spent on staff salaries.
Dejene Girma (PhD) Dilla University Department of Mathematics
 Statistics: What’s What?

Descriptive objectives/ Comparative objectives/

research questions: hypotheses

Descriptive statistics Inferential Statistics

summarize data are tools that indicate how much
confidence we can have when we
generalize from a sample to a
population.
Dejene Girma (PhD) Dilla University Department of Mathematics
 Descriptive Statistics

Can be applied to any measurements (quantitative or

qualitative)
Offers a summary/ overview/ description of data.

Does not explain or interpret.

Dejene Girma (PhD) Dilla University Department of Mathematics

 Descriptive Statistics

Number Variability
Frequency Count Variance and standard
Percentage deviation
Deciles and quartiles Graphs
Measures of Central Normal Curve
Tendency (Mean,
Midpoint, Mode)

Dejene Girma (PhD) Dilla University Department of Mathematics

 Means of Central Tendency

Averages
 Mode: most frequently occurring value in a
distribution (any scale, most unstable)
 Median: midpoint in the distribution below which
half of the cases reside (ordinal and above)
 Mean: arithmetic average- the sum of all values in a
distribution divided by the number of cases (interval
or ratio)

Dejene Girma (PhD) Dilla University Department of Mathematics

 Median (Mid-point)

Example (11 test scores)

61, 61, 72, 77, 80, 81, 82, 85, 89, 90, 92
The median is 81 (half of the scores fall above 81, and half
below)
Example (6 scores)
3, 3, 7, 10, 12, 15
Even number of scores= Median is half-way between these scores
Sum the middle scores (7+10=17) and divide by 2
17/2= 8.5

Dejene Girma (PhD) Dilla University Department of Mathematics

 Median

Insensitive to extremes
3, 3, 7, 10, 12, 15, 200
Mean: Arithmetic Average
Mean is half the sum of a set of values:
Scores: 5, 6, 7, 10, 12, 15
Sum: 55
Number of scores: 6
Computation of Mean: 55/6= 9.17

Dejene Girma (PhD) Dilla University Department of Mathematics

 Mean
Influenced by extremes
Only appropriate with interval or ration data
Is this four-point scale ordinal or interval?
1= Strongly Agree 3=Disagree
2=Agree 4=Strongly Disagree
Mode: Frequency
Mode is the most frequently occurring value in a set.
Best used for nominal data.

Dejene Girma (PhD) Dilla University Department of Mathematics

 Shapes of Distribution

Normal Curve (aka Bell Curve)

Repeated sampling of a population should result in a
“normal” distribution- clustering of values around a central
tendency.
In a symmetrical distribution, median, mode and mean all
fall at the same point

Dejene Girma (PhD) Dilla University Department of Mathematics

Dejene Girma (PhD) Dilla University Department of Mathematics
 Normal Curve

Dejene Girma (PhD) Dilla University Department of Mathematics

 Distribution: Skewness

Skewed to the right (positive) or left (negative)

An extremely hard test that results in a lot of low grades
will be skewed to the right:

Dejene Girma (PhD) Dilla University Department of Mathematics

 Positive

the mode is smaller than the median, which is smaller than

the mean.
This relationship exists because the mode is the point on the
x-axis corresponding to the highest point, that is the score
with greatest value, or frequency.
The median is the point on the x-axis that cuts the
distribution in half, such that 50% of the area falls on each
side.

Dejene Girma (PhD) Dilla University Department of Mathematics

 Negative

An extremely easy test will result in a lot of high grades,

and will skew to the left (negative)

The order of the measures of central tendency would be the

opposite of the positively skewed distribution, with the
mean being smaller than the median, which is smaller than
the
Dejene Girma mode.
(PhD) Dilla University Department of Mathematics
 Variability

Variability is the differences among scores- shows how

subjects vary:
 Dispersion: extent of scatter around the “average”
 Range: highest and lowest scores in a distribution
 Variance and standard deviation: spread of scores in
a distribution. The greater the scatter, the larger
the variance
Interval or ratio level data
Standard deviation: how much subjects differ from the
mean of their group

Dejene Girma (PhD) Dilla University Department of Mathematics

 Standard Deviation

Measures how much subjects differ from the mean of

their group
The more spread out the subjects are around the
mean, the larger the standard deviation
Sensitive to extremes or “outliers”

Dejene Girma (PhD) Dilla University Department of Mathematics

 Standard Deviation: 66, 95, 99%

Dejene Girma (PhD) Dilla University Department of Mathematics

 Inferential Statistics

Allows for comparisons across variables

i.e. For Example
 Is there a relation between one’s occupation and their

reason for using the public library?

 Is there a relation between Family income and

Students Mathematics achievement ?

Dejene Girma (PhD) Dilla University Department of Mathematics

 Levels of significance

Hypothesis Testing
The level of significance is the predetermined level at which a null
hypothesis is not supported.
The most common level is p < .05
[eg. Ho : No significance relation between Family income and Mathematics
Achievement]
 P =probability

 < = less than (> = more than)

Dejene Girma (PhD) Dilla University Department of Mathematics

 Error Type

Type I error Type II error

Reject the null Fail to reject the null
hypothesis when it is hypothesis when it is
really true really false

Dejene Girma (PhD) Dilla University Department of Mathematics

 Probability

By using inferential statistics to make decisions, we

can report the probability that we have made a Type I
error (indicated by the p value we report)
By reporting the p value, we alert readers to the odds
that we were incorrect when we decided to reject the
null hypothesis

Dejene Girma (PhD) Dilla University Department of Mathematics

 Particular Tests

Chi-square test of independence: two variables (nominal

and nominal, nominal and ordinal, or ordinal and
ordinal)
 Affected by number of cells, number of cases
 2-tailed distribution= null hypothesis
 1-tailed distribution= directional hypothesis
 Cramer’s V, Phi

Dejene Girma (PhD) Dilla University Department of Mathematics

 Inferential Statistics (2)

Correlation—the extent to which two variables are

related across a group of subjects
 Pearson r
 It can range from -1.00 to 1.00
 -1.00 is a perfect inverse relationship—the strongest possible inverse
relationship
 0.00 indicates the complete absence of a relationship
 1.00 is a perfect positive relationship—the strongest possible direct
relationship
 The closer a value is to 0.00, the weaker the relationship
 The closer a value is to -1.00 or +1.00, the stronger it is

Dejene Girma (PhD) Dilla University Department of Mathematics

 …..Correlation

Spearman rho
In statistics, Spearman's rank correlation coefficient or Spearman's ρ,
named after Charles Spearman and often denoted by the Greek letter \
rho or as r(s), is a nonparametric measure of rank correlation.
It assesses how well the relationship between two variables can be
described using a monotonic function
ρ = Spearman's rank correlation coefficient
di = difference between the two ranks of each observation
n = number of observations

Dejene Girma (PhD) Dilla University Department of Mathematics

 More tests

t-test
 Test the difference between two sample means for significance
 pretest to posttest

It is used Relates to research design

Eg. To see the effect of Problem Solving Approach
teaching for students math achievement (pre test
Post test and compare the mean)
Perhaps used for information literacy instruction
Analysis of variance
Regression analysis (including step-wise regression)
Dejene Girma (PhD) Dilla University Department of Mathematics
 More tests

ANOVA
Analysis of variance (ANOVA) tests the difference(s) among two
or more means
It can be used to test the difference between two means

So use t-test or ANOVA?

KEY: ANOVA also can be used to test the difference among

more than two means in a single test—which cannot be done

with a t test
Dejene Girma (PhD) Dilla University Department of Mathematics
 More tests

While correlation and regression both indicate association

between variables, correlation studies assess the strength of that
association
Regression analysis, which examines the association from a
different perspective, yields an equation that uses one variable to
explain the variation in another variable.
Regression is used to predict the value of one variable by
knowing the value of another variable

Dejene Girma (PhD) Dilla University Department of Mathematics

 more tests

Multiple regression examines the relationship between a dependent variable

(changes in response to the change the researcher makes to the independent

variable) and two or more independent variables (manipulated variables)
Stepwise multiple regression predicts the value of a dependent variable using

independent variables, and it also examines the influence, or relative

importance, of each independent variable on the dependent variable

Dejene Girma (PhD) Dilla University Department of Mathematics

 Parametric and Nonparametric statistics

Parametric statistical tests generally require interval or

ratio level data and assume that the scores were drawn
from a normally distributed population or that both sets of
scores were drawn from populations with the same
variance or spread of scores
Nonparametric methods do not make assumptions about
the shape of the population distribution. These are typically
less powerful and often need large samples

Dejene Girma (PhD) Dilla University Department of Mathematics

 Selecting an Appropriate Statistical
Test

The appropriate measurement scale(s) to use Is

intent to characterize respondents (descriptive
statistics) or draw inferences to population
(inferential statistics)
The level of significance used and focusing on one-
or two-tailed distribution
Whether the mean or median better characterize
the dataset
Whether the population is normal
Dejene Girma (PhD) Dilla University Department of Mathematics
 Selecting an Appropriate Statistical
Test

The number of independent (experimental or predicator

variables that evaluators manipulate and that
presumably change) and dependent (influenced by the
independent variable(s))
Uses parametric or nonparametric statistics
Willing to risk a type I or type II errors
I: possibility of rejecting a true null hypothesis
II: possibility of accepting the null hypothesis when it
is false

Dejene Girma (PhD) Dilla University Department of Mathematics

 s s o n
f Le
d o
En

Dejene Girma (PhD) Dilla University Department of Mathematics