SAPP Module 5: Hypothesis Testing  Applicable when probability sampling is used (note that Inferential

Statistics is founded on the Theories of Probability).

Lesson 1: Preliminaries to Hypothesis Testing  Questionable results when non-probability sampling is used.

Introduction to Inferential Statistics

In PSYCH101 (Descriptive Statistics), we discussed the general philosophy of formal
statistical inference. Statistical Inference consists of those methods by which one
makes inferences or generalizations about a population (Walpole et al., 2012).

Purpose of Inferential Statistics

Statistics (Inferential) help us in making decisions, establishing numerical
benchmarks, and many more. This includes forming reliable conclusions; and the
decision making starts with the testing of hypothesis. (Walpole et al., 2012).
Inferential Statistics is also called as Inductive Statistics (APA Dictionary of
Psychology, 2023).
Appropriateness of Inferential Statistical Tools In Hypothesis Testing
Recall. Population v. Samples  The level/scale of measurement as well as the type of variable are
important indicators in the choice of statistical inferential procedures,
Population especially in hypothesis testing.
- complete set of individuals  Parametric tests have more stringent assumptions before they can be used:
- Parameters - characteristics from the population data must be at least interval (i.e., quantitative) and the distribution of the
Sample data must be normal or approximately normal (recall the Central Limit
- subset of the population Theorem).
- Statistic - characteristics from the sample
Hypothesis Testing: Parametric and Non-Parametric Tests
Statistical Inference is the act of generalizing from a sample to a population with a  Parametric Tests are generally more powerful and robust than
calculated degree of certainty. That is, we want to learn about population nonparametric tests.
parameters...but we can only calculate sample statistics.  Parametric Tests relies on the assumption that the random samples are
selected from normal (or approximately normal provided the sample size is
Purpose of Inferential Statistics for Psychologists large) populations (Walpole et al., 2012).
Recall that Inferential statistics provide researchers with the tools to make
 Nonparametric tests are used in cases where parametric tests are NOT
inferences about the meaning of the results (Walter, 2020). Specifically, they allow
appropriate. They are about 95% as powerful as parametric tests (Chin &
researchers to generalize from the sample they used in their research, to the greater
Lee, 2008).
population, which the sample represents. KEEP IN MIND that psychologists, like
 Some common situations for using nonparametric tests are when:
other scientists, rely on relatively small samples to try to understand populations
o The distribution is not normal (hence, skewed);
(Walter, 2020).
o The distribution is not known (hence, distribution-free);
o The sample size is too small (less than 30) to assume a normal
Assumptions of Inferential Statistics
distribution (see Central Limit Theorem).
 Allows conclusions about the population to be generated from sample
o There are extreme values (outliers).
data.
o The variables are in nominal or ordinal scale (Chin & Lee, 2008)
Some Common Research Methodological Problems
A. BIAS - A methodological problem resulting from a selection process that is not
random, i.e., not due to chance.
1. Observer Bias
– some error or distortion (either intentional or unintentional) in the
perception or description of the data by the observer.
– the observer imposes their own expectations onto the subjects.
– occurs when researchers’ preferences or expectations influence the
outcome of their research. In these cases, researchers see what they
want to see rather than what is actually there.
2. Subject Bias
– some error or distortion of the measurement by the study subject.
– Research subjects’ expectations can affect and change the subjects’ Lesson 3: Steps in Hypothesis Testing
behavior, resulting in subject bias.
– Such a bias can manifest itself in two ways: Hypothesis testing is a decision-making process for evaluating claims about the
o A placebo effect is the effect on a subject receiving a fake drug or population. In hypothesis testing, the researcher must define the population
treatment. understudy, State the particular hypothesis that will be investigated, give the
o The social desirability bias is the tendency of some research subjects significance level, select a sample from the population, collect the data, perform the
to describe themselves in socially approved ways. calculations required for the statistical test and reach occlusion.
3. Instrument Bias
– some error or distortion from faulty mechanical equipment, such as an Three methods used to test hypothesis are:
uncalibrated scale that gives falsely low weight. 1. The traditional method
– A survey instrument or questionnaire that does NOT capture data to 2. The p-value method
properly reflect the views of the respondent group contains an 3. The confidence interval method
instrumentation bias.
B. ERROR - is the difference between the true value of a measurement and the STEPS in HYPOTHESIS TESTING
recorded value of a measure.
1. Random Error STEP 1. State the hypotheses.
– a wrong result due to chance.
– known as variability, random variation, or ‘noise in the system’ (i.e., Every hypothesis testing situation begins with the statement of a hypothesis. a
random irregularity we find in any real-life data). Statistical Hypothesis is an assertion or conjecture concerning one or more
– corresponds to imprecision populations (Walpole, et.al., 2012).
2. Systematic Error
– a wrong result due to interference from bias. Null Hypothesis (Ho) - a statistical hypothesis that states that there is no difference
– refers to deviations that are not due to chance alone. between a parameter and a specific value, or that there is no difference between
– corresponds to inaccuracy two parameters.

Example:
An instrument that is improperly calibrated so that it consistently overestimates (or
underestimates) the measurements.
This is the statement of “zero” difference/relationship. This assumes that there is no
difference/relationship between two means or variables being compared. This is the
hypothesis that is hoped to be rejected.

Alternative Hypothesis (Ha) - a statistical hypothesis that states the existence of a

difference between a parameter and specific value, or states that there is a
difference between two parameters.

This is the statement that assumes there is a significant difference between the two
means or variables under investigation. This is the hypothesis that is accepted when
the null hypothesis is rejected.

Lesson 4: Level of Significance and Critical Values

Level of Significance and Power of a Test (McCarthy, 2021)
 Level of Significance (⍺) is the probability of rejecting the null hypothesis while
it is true.
 Power (1-β) is the probability of rejecting the null hypothesis while it is false.
 Level of Significance (⍺) is, thus, the probability of Type I error, whereas
 Power (1-β) is the probability of avoiding a Type II error. Suppose there is a test for a particular disease.
 If all other things are held constant, then as level of significance (⍺) increases,  If the disease really exists and is diagnosed early, it can be successfully
so does the power of the test (1-β) increases (Bullard, 2023). treated.
 This is because a larger ⍺ means a larger rejection region for the test and thus a  If it is not diagnosed and not treated, the person will become severely
greater probability of rejecting the null hypothesis. disabled.
 This translates to a more powerful test.  If a person is erroneously diagnosed as having the disease and treated, no
 The price of this increased power is that as α goes up, so does the probability of physical damage is done.
a Type I error should the null hypothesis in fact be true (Bullard, 2023).
 An alpha level of 0.05 IS REASONABLE and is WIDELY ACCEPTED by all scientists
and researchers (Amparo, 2024).
 THE NULL HYPOTHESIS CAN BE REJECTED IF THERE IS LESS THAN 0.05
PROBABILITY OF COMMITTING A TYPE I ERROR (P < 0.05).

The next parts are according to Bolen (2019):

 The smaller the value of ⍺, the surer we are that we are NOT making an error if we
end up rejecting the null hypothesis. Thus, a smaller ⍺ leads to a higher amount of
“evidence” needed before we reject the null hypothesis in favor of the alternative
hypothesis.

The critical or rejection region is the range of values of the test value that indicates
that there is a significant difference and that the null hypothesis should be rejected.

The non-critical or non-rejection region is the range of values of the test value that
indicates that the difference was probably due to chance and that the null
hypothesis should not be rejected

One-tailed test
- indicates that the null hypothesis should be rejected when the test value is in the
critical region on one side of the mean.
- Can be either right-tailed or left-tailed, depending on the direction of the
The level of significance ⍺ of a test is the probability of rejecting the null hypothesis inequality of the alternative hypothesis (>, <) .
when it is true. In short, ⍺ is the maximum probability of committing a Type I error.
Hence, P (Type I error) = ⍺ Two-tailed test
- the null hypothesis should be rejected when the test value is in either of the two
Commonly used values of ⍺ = 0.05, 0.01. critical regions.
- When the alternative hypothesis has the not equal sign (≠) statistical hypothesis.
 Lesson 5 - Traditional and P-value Methods in Hypothesis Testing

I. TRADITIONAL METHOD

Procedure: Solving Hypothesis-Testing Problems ( Traditional Method)

State the hypotheses and identify the claim.

Find the critical value (s) from the appropriate table.
Compute the test value.
Make the decision to reject or not reject the null hypothesis.
Summarize the results.
II. P-Value Method for Hypothesis Testing

The P-Value ( or probability value ) is the probability of getting a sample statistic

( such as the mean) or a more extreme sample statistic in the direction of the
alternative hypothesis when the null hypothesis is true.

Procedure: Solving Hypothesis-Testing Problems ( P-Value Method)

State the hypotheses and identify the claim.

Compute the test value and find the P-value.
Make the decision
Summarize the results.
STEPS IN USING THE STATISTICAL TOOL BOX (ANDROID PHONE)

1. Choose Statistical tests

2. Then choose Student's t-tests (One sample )
3. Encode each value separated by commas. ex. 13500,16000,12000,
4. Enter the population mean under "Enter Ho of assumed mean"
5. Press perform test.
Default level of significance = 0.05
Dependent Samples T-test Jamovi Flowchart