Definition

Inferential statistics aims at drawing conclusions about

a population from a sample. This is generally done through random sampling that leads to central tendency of a distribution. Inferential statistics offers

more powerful analyses to be performed on a survey data. This can

include associations between variables, how well a sample represents a

certain population, and cause-and-effect relationships. (Freedman, 2009;

Bandyopadhyay & Forster, 2011). An example of inferential

statistics is shown in the following figure

Figure 1: The Relation

between Potatoes Distribution and Week Days

Importance

Cox

(2006) points out that inferential statistics are important as it:

1.

Is effective in predicting and inferring

a much larger data or population.

2. Deeply analyzes

the statistical data and observations.

3.

Studies more details

such as hypothesis tests and confidence interval.

4. Covers a

wide range of data as the measures in inferential statistics are not always

exact numbers.

5.

Places the educated predictions

and guesses on the basis of the parameters of the given population, it

does not matter how big the population is.

6.

Plays an important role in examining

the relationships between variables within a sample, and then make

generalizations or predictions about how those variables will relate within a

larger population.

1. The Main Types of Inferential

Statistics

There are two main types of inferential

statistics:

1) Confidence Interval:

It refers to the form of an interval that provides a range

for the parameter of given population.

2) Hypothesis Test (Test of Significance):

Bickel & Kjell (2001) believe that to conduct inferential

statistics, it is important and necessary to conduct test of significance in

order to know whether results can be generalized to a larger population. Common

tests of significance include the Chi-square and T-test. These determine

the probability that the results of statistical analysis are representative of

the population.

Besides, Freedman (2010) states that there are other

techniques that are used to examine the relationships between variables, and

thereby to create inferential statistics. They include linear regression analyses, logistic regression analyses, ANOVA, correlation analyses, structural equation modeling, and survival analysis.

Linear

Regression and Multiple Linear Regressions

Newsom (2010)

points out that linear regression is a statistical technique that is used to

learn more about the relationship between an independent (predictor) variable

and a dependent (criterion) variable. If the independent variable is more than

one, this is referred to as multiple linear regressions.

R-Square

R-square, also known as the coefficient of determination, is a commonly used

statistic to evaluate the model fit of a regression equation. That is, how good

are all of independent variables at predicting dependent variable? The value of

R-square ranges from 0.0 to 1.0 and can be multiplied by 100 to obtain a

percentage of variance explained.

Logistic

Regression

Logistic regression is a method for modeling a binary

response variable, which takes values 0 and 1. The dependent variable is always

binary whereas the independent, or predictor, variables can be either numerical

or categorical.

Analysis of

Variance (ANOVA)

It clarifies the

significant differences between means. It is of different models: one-way

between groups ANOVA, one-way repeated measures ANOVA, two-way between groups ANOVA, two-way repeated measures ANOVA

Correlation Analysis

Correlation analysis verifies the relationship between two

variables where a high, correlation means that two or more variables have a

strong relationship with each other whereas a low correlation means that the

variables are hardly related.

Structural Equation Modeling

Structural equation modeling uses some statistical techniques

that allow a set of relationships between one or more independent variables and

one or more dependent variables to be examined. Both variables could be either

continuous or isolated and can be either factors or measured variables.