Definition of Key Concepts in Statistics

What is Statistics?

●​ Statistics is a study that deals with the collection, organization, analysis,

interpretation, and presentation of data to understand or solve problems.
It helps in making informed decisions based on numerical/empirical
evidence.

Types of Statistics:

1.​ Descriptive Statistics – Summarizes and presents given data in a

meaningful way, which can be a representation of the entire population
or a sample.
○​ Measures of Central Tendency: Mean, Median, Mode
○​ Measures of Dispersion: Range, Variance, Standard Deviation
2.​ Inferential Statistics – Analyzes sample data to make predictions or
inferences about a larger population. It involves drawing a conclusion.
○​ Hypothesis Testing
○​ Confidence Intervals
○​ Regression Analysis

Identification of Data Types and Measurement Scales

Types of Data

1.​ Categorical (Qualitative Data) - Represents labels, groups, or categories. It

does not have numerical meaning but is used to classify observations.
○​ Nominal: Categories with no specific order (e.g., gender, colors, car
brands).
○​ Ordinal: Categories with a meaningful order/rank but unequal
intervals (e.g., customer satisfaction ratings: Poor, Fair, Good,
Excellent).
 2.​ Scale (Quantitative Data) - Consists of numerical values that represent
measurable quantities like height, weight, or temperature.
○​ Discrete: Integers, whole numbers, or counting numbers (number of
employees, number of products sold, allowance).
○​ Continuous: Contains fractions and decimals (e.g., weight, height,
temperature).

Scale Measurement:

●​ Ratio: There is a real zero.

●​ Interval: No real zero.
●​ Ordinal: Ranking
●​ Nominal: Numbers

Problems for Descriptive Statistics

Measures of Central Tendency - Indicates the center of the data set. Subject to
outliers. (Vulnerability of data using mean to compare).

1.​ Mean (Average) - The sum of all values divided by the number of values.
○​ Formula: Mean= Summation of X/N
○​ Best for: Normally distributed data without extreme values.
2.​ Median - The middle value when data is arranged in ascending order.
○​ Best for: Skewed data or data with outliers.
○​ More stable than the mean because it is not susceptible to outliers.
3.​ Mode - The most frequently occurring value in a dataset.
○​ Example: In 2, 3, 3, 4, 5, 5, 5, 6, the mode is 5.
○​ Best for: Categorical data or identifying common values in a
dataset.

PARAMETER - population
STATISTICS(n) - SAMPLE
Measures of Variation - Refers to how spread out or dispersed the data points
are in a dataset from the central tendency. It helps measure the consistency or
inconsistency of values and provides insight into the reliability of data.

1.​ Range - Difference between the highest and lowest values.

○​ Formula: Range= Max−Min
○​ Example: Highest salary is ₱80,000, lowest is ₱20,000 → Range =
₱60,000.
○​ Limitation: Only considers two values and ignores distribution.
2.​ Variance
○​ Measures the average squared difference from the mean.
○​ High Variance: More spread out data.
○​ Low Variance: Data points are close to the mean.

3.​ Standard Deviation

○​ The square root of variance; shows the average distance of data
points from the mean.
 4.​ Interquartile Range (IQR) - The Interquartile Range (IQR) is a measure of
statistical dispersion that shows the spread of the middle 50% of a dataset/
divided intro 4 equal parts(25%). It is useful for identifying variability while
reducing the impact of outliers.

IQR 5 VALUES SUMMARY:

○​ Maximum - X largest
○​ Minimum - X smallest
○​ Q3
○​ Q4
○​ Q1.

How to Calculate IQR:

1.​ Arrange the data in ascending order.

2.​ Find the first quartile (Q1) – the median of the lower half (25th percentile).
3.​ Find the third quartile (Q3) – the median of the upper half (75th
percentile).
4.​ Compute the IQR using the formula: IQR=Q3−Q1

​ N(big N)usually refers to a population size,

n(small n) refers to a sample size.

Example:

Consider the dataset: 5, 7, 9, 12, 15, 18, 22, 26, 30

■​ Q1 (25th percentile): 9
■​ Q3 (75th percentile): 22
■​ IQR: 22 - 9 = 13

Formulas:
 Answer:

NOTE: If 5 ra ang sample (n), automatically apply the IQR.

= Xsmallest, Q1,Q2,Q3,Xlargest

Formulation of Research Question, Hypothesis, Type of Variables

Variables

●​ Independent Variable (X): The variable that is manipulated or

categorized.
●​ Dependent Variable (Y): The outcome that is measured.

Formulating Hypotheses

1.​ Null Hypothesis (H₀): No significant relationship/difference.

2.​ Alternative Hypothesis (H₁): A significant relationship/difference exists.

Identifying the Appropriate Test

Parametric vs. Non-Parametric Tests

●​ Parametric Tests: Parametric tests are used to answer questions involving

magnitude, effects, or influence of a variable to another variable, or test of
 differences (i.e., comparison of means). Used for numerical data when
making inferences about a population (e.g., t-test).
●​ Non-Parametric Tests: Non-parametric tests on the other hand, are tests for
categorical data (i.e., nominal vs. nominal, nominal vs. ordinal, ordinal vs.
ordinal) Used for categorical data (e.g., chi-square test).

Chi-square test or test of independence is a common non-parametric test. Key

concepts for chi-square test are:

●​ Contingency table (e.g., 2 x 2, 3 x 5)

●​ Observed frequency (fo)- actual counts
●​ Expected frequency (fe) - calculated as column total*row total/Grand total
●​ Grand total (N) - the sum of the column totals or the row totals
●​ Degrees of freedom (df) - calculated as no. of rows - 1 multiplied by no. columns
-1
●​ the probability - to be compared with the level of significance (to conclude
whether the variables are independent or not)
●​ the Chi-square value

Limitations of Chi-square test

* All participants measured must be independent, meaning that an individual cannot fit
in more than one category. Example, in our sample exercise, we tested the association
between gender (male & female) and pet preference (dog & cat). A male respondent
must not prefer both dog and cat, a female respondent must not prefer both dog and
cat. Otherwise, the chi-square test will not be appropriate.
* The data must be a frequency data, meaning, actual counts.
* No more than 20% of the expected frequencies should be less than 5 in a contingency
table and all individual expected counts are 1 or greater. If this rule is violated, the
Chi-Square test may not be valid, and an alternative test (such as Fisher’s Exact Test)
should be considered. See examples for this rule in the next material.
*If the sample size is less than 50, Chi-square test is not appropriate. Instead, Fisher's
Exact Test is recommended.

Tests of Relationship

1.​ Test of Independence or Association (Chi-Square Test)

○​ Determines if two categorical variables are related.
○​ Formula: χ2=∑(O−E)2E\chi^2 = \sum \frac{(O - E)^2}{E}χ2=∑E(O−E)2​
where O = Observed frequency, E = Expected frequency.
 2.​ Test of Relationship for Numerical Variables (Correlation Test)
○​ Measures the strength and direction of a relationship between two
numerical variables.
○​ Correlation Coefficient (r):

Determining the results of two numerical variables.

■​ 0 - 0.19: Very weak

■​ 0.20 - 0.39: Weak
■​ 0.40 - 0.59: Moderate
■​ 0.60 - 0.79: Strong
■​ 0.80 - 1.00: Very strong
3.​ T-test (Comparison of Means)
○​ Independent Sample T-test: Compares two different groups.
○​ Paired Sample T-test: Compares the same group at two different
times.
○​ One-tailed T-test: Tests if one mean is greater or less than another.
○​ Two-tailed T-test: Tests if means are different in any direction.

Identifying One-Tailed or Two-Tailed Test

●​ One-tailed test: Tests if a parameter is greater or less than a value.

●​ Two-tailed test: Tests if a parameter is different from a value, regardless of
direction.

In detail:

●​ One-tailed T-test:
○​ Used when the research hypothesis suggests a specific direction (e.g.,
Method B is greater than Method A).
○​ Rejection Region:
■​ If Right-tailed: The rejection region is on the right side of the
distribution (for hypotheses suggesting an increase or greater than).
■​ If Left-tailed: The rejection region is on the left side (for hypotheses
suggesting a decrease or less than).
●​ Two-tailed T-test:
 ○​ Used when the hypothesis does not suggest a specific direction (e.g.,
there is a difference but unsure whether greater or lesser).
○​ Rejection Region: Both ends of the distribution.

Level of Significance (Alpha):

●​ It is the threshold for rejecting the null hypothesis.

●​ Standard values are 0.05 or 0.01, depending on the desired confidence level

Interpreting Results of Hypothesis Testing

1.​ Statistical Hypothesis - Rejection vs. Non-Rejection

○​ Reject H₀: If the sample evidence strongly contradicts it.
○​ Do not reject H₀: If evidence is insufficient to reject it.
2.​ Level of Significance (α)
○​ Common values: 0.05 (5%) or 0.01 (1%).
○​ If p-value < α, reject H₀.
○​ If p-value ≥ α, do not reject H₀.
3.​ Decision Rule for T-Test
○​ For Positive T-stat or P-value:
i.​ If T-stat > T-critical or P-value > Alpha → Reject the Null Hypothesis
ii.​ If T-stat ≤ T-critical or P-value ≤ Alpha → Do Not Reject the Null
Hypothesis
○​ For Negative T-stat or P-value:
i.​
ii.​ If -T-stat < -T-critical or P-value < Alpha → Reject the Null Hypothesis
iii.​ If -T-stat ≥ -T-critical or P-value ≥ Alpha → Do Not Reject the Null
Hypothesis
4.​ Conclusion
○​ If the test shows significance, the alternative hypothesis is
accepted.
○​ If not, the null hypothesis is not rejected.

OTHER IMPORTANT NOT IN THE POINTERS:

FREQUENCY
In statistics, frequency is the number of times a value appears in a set of data. It can
also be described as the number of times an event occurs in an experiment or study.

How is frequency expressed?

• The number of times a value occurs
• The number of times a value occurs compared to the total number of values
• Ways to express relative frequencies

For example, in our exercise:

* The number of times male or females appears in our data set.
* The number of times job experience of below 5 years or 5 years and above occur in
our data set.
* The number of times very good, good, fair, or poor job performance occur in our data
set.
Other examples:
• Counting how many students scored a particular mark on a test
• Counting how many people own dogs in a survey

Importance of Frequency in Statistics

In statistics, "frequency" is important because it tells you how often a specific value
appears within a dataset, allowing you to understand the distribution of data, identify
patterns, and visualize how frequently certain values occur, which is crucial for
interpreting and analyzing data effectively; essentially, it provides a clear picture of
which data points are most common and which are less frequent within a set.

Key points about frequency in statistics:

• Frequency distribution: A table or graph that displays the frequency of each unique
value in a dataset, providing a visual representation of data distribution.
• Frequency table: A table that shows how often an event happened
• Understanding patterns: By looking at the frequency of different values, researchers
can identify trends, outliers, and areas of concentration within the data.
• Data Analysis Applications: Frequency analysis is used in various fields like market
research, quality control, social science, and healthcare to analyze customer behavior,
product defects, demographic trends, and more.
• Relative Frequency: The proportion of times a value appears within a dataset, often
expressed as a percentage, which allows for comparison between different data sets
even if the total sample sizes vary.

Elements in Hypothesis Testing (Steps) for T-tests

1. Ho (null hypothesis)
 H1 (alternative hypothesis)
2. α = level of significance
3. T statistic (T-value) = calculated T
4. Rejection region = t-critical or tabulated t
5. Compare: T > t or -T < -t ; P < α
6. Conclusion: Reject or Do not reject Ho
7. Insights: Interpretation of results, added with related literature (APA style, 7th edition)

Explore the following concepts:

1. statistical inference
2. tests of hypothesis
3. statistical hypothesis - rejection vs. non-rejection of hypothesis
4. level of significance (alpha)
5. one-tailed test vs. two-tailed test
6. independent variable and dependent variable in a schema
7. statement of the problem