Psychological Statistics Notes

Psychological Statistics with Renz Marion Gavino

Interesting Statistics

  • As of 2021, the average life expectancy in the Philippines is approximately 71 years.

  • 1 in 5 Filipinos experiences mental health issues in their lifetime.

  • Only 1 in 5 individuals with mental health problems in the Philippines seeks treatment.

  • Over 92% of the population now has access to clean and safe drinking water.

  • A new advertisement for a popular ice cream brand led to a 30% increase in ice cream sales over the next three months, suggesting a positive impact on consumer demand.

  • The statement that more churches in a city lead to more crime is presented with a caution against assuming causation.

  • There are 75% more interracial marriages occurring this year than 25 years ago, suggesting increased societal acceptance of interracial marriages.

  • It is noted that not everything read online, even with the prompt "psychology says…", is correct.

Statistical Procedures

  • Learning Outcome: Equip students with the tools and techniques needed to analyze, interpret, and draw meaningful conclusions from data, enabling them to make informed decisions and contribute to evidence-based practices in the field of psychology.

  • Statistics is the branch of mathematics that deals with the collection, analysis, interpretation, presentation, and organization of data.

  • It involves using various methods to summarize and draw conclusions from data sets, helping to identify patterns, relationships, or trends.

  • Statistics is the language of science and data.

Population

  • A population is the set of all the individuals of interest in a particular study.

  • A researcher may want to know what factors are associated with academic dishonesty among college students.

  • A researcher may want to examine the amount of time spent in the bathroom for men compared to women.

  • Because populations tend to be very large, it is usually impossible for a researcher to examine every individual in the population of interest.

Sample

  • A sample is a set of individuals selected from a population, usually intended to represent the population in a research study.

  • Samples can vary in size. For example, one study might examine a sample of only 10 students in a graduate program, and another study might use a sample of more than 10,000 people who take a specific cholesterol medication.

Variables

  • Variables are characteristics or conditions that change or have different values for different individuals.

  • Independent Variable: The variable in an experiment that is systematically manipulated by the investigator.

  • Dependent Variable: The variable in an experiment that the investigator measures to determine the effect of the independent variable.

Data

  • A datum (singular) is a single measurement or observation and is commonly called a score or raw score.

  • Data (plural) are measurements or observations.

  • A data set is a collection of measurements or observations.

  • Discrete Data: Data that can take only specific values within a range. These values are countable and usually represent whole numbers.
    Operations: Can be counted but not divided into smaller parts.
    Values: Can only take whole numbers or integers.
    Examples: Number of children in a family, number of cars in a parking lot, number of books on a shelf.

  • Continuous Data: Data that can take any value within a range and is measurable. It can take on an infinite number of values within a given interval.
    Operations: Can be measured and expressed with infinite precision (e.g., 2.34567).
    Values: Can take any value within a given range, including decimals or fractions.
    Examples: Height of a person (e.g., 5.6 cm, 5.75 cm), weight, temperature (e.g., 20.5°C), time taken to complete a task.

  • Categorical Data: Data that represents categories or labels, which are typically non-numeric. It involves grouping items based on characteristics. Qualitative data representing different categories or groups.
    Operations: Can be used for classification or labeling, but no meaningful arithmetic operations like addition or subtraction.
    Values: Can take values representing different categories, but no inherent order (nominal) or with a natural ranking (ordinal).
    Examples: Gender (Male/Female), hair color (Black/Blonde/Red), marital status (Single/Married), favorite fruit (Apple/Orange/Banana).

  • Countable: You can count it in whole numbers (like 1, 2, 3).

  • Measurable: You can measure it and get values with decimals (like 5.25, 3.5).

  • Data Measurement Criteria:

    • Is the data countable or measurable?

      • Discrete: Countable

      • Continuous: Measurable

      • Categorical: Neither

    • Can the data take any value, including fractions or decimals?

      • Discrete: No

      • Continuous: Yes

      • Categorical: No

    • Does the data represent categories or labels?

      • Discrete: No

      • Continuous: No

      • Categorical: Yes

Scales of Measurement

  • Ratio scales have equal intervals between adjacent scores on the scale and an absolute zero.

  • Interval scales have equal intervals between adjacent scores but do not have an absolute zero.

  • Ordinal scales have some sort of order to the categories (e.g. in terms of magnitude) but the intervals between adjacent points on the scale are not necessarily equal.

  • Nominal scales consist of categories that are not ordered in any particular way.

SCALE

NOMINAL

ORDINAL

INTERVAL

RATIO

DEFINITION

Data categorized into distinct groups without any order. Qualitative

Data categorized in a specific order or ranking.

Data where the difference between values is meaningful, but there is no true zero point.

Data with meaningful differences and a true zero point.

DATA REPRESENTATION

Categories or labels. (No numerical value)

Ranks or order of data (e.g., 1st, 2nd, 3rd)

Numbers where intervals between values are equal, but zero doesn’t mean "none."

Numbers with a true zero value (e.g., zero means none).

MATHEMATICAL OPERATIONS

No meaningful math operations.

Can compare rankings (greater than, less than).

Can add and subtract, but ratios do not make sense.

Can perform all arithmetic operations (addition, subtraction, multiplication, and division).

EXAMPLES

Gender (Male/Female), Eye color (Blue/Green)

Education level (High School, College, Graduate)

Temperature (Celsius/Fahrenheit), IQ score

Height, Weight, Age, Salary

Criteria for Scales:

CRITERIA

NOMINAL

ORDINAL

INTERVAL

RATIO

Does the scale involve categories or labels?

Is there a meaningful order or ranking among categories?

Are the differences between values equal and meaningful?

Does the scale have a true zero point (i.e., zero means "none")?

Can you perform all arithmetic operations (addition, subtraction, multiplication, division)?

  • Examples of Scales:

    • Nominal: Favorite color (red, blue, green), Car brands (Toyota, Ford, BMW), Pets (dog, cat, hamster), Fruit types (apple, banana, orange).

    • Ordinal: Movie ratings (1 star, 2 stars, 3 stars, 4 stars, 5 stars), Spicy food levels (mild, medium, hot, extra hot), Clothing sizes (small, medium, large), Ranking of sports teams (1st place, 2nd place, 3rd place).

    • Interval: Temperature (Celsius/Fahrenheit) (10 °C, 20 °C, 30 °C), Time on a clock (2:00 PM, 3:00 PM, 4:00 PM), Calendar years (2020, 2021, 2022).

    • Ratio: Height (0 cm, 100 cm, 200 cm), Weight (0 kg, 50 kg, 100 kg), Money ($0, $10, $100), Distance (0 meters, 5 meters, 10 meters).

  • Interval vs Ratio:

    • A 1-10 effectiveness scale lacks an absolute zero and is interval data.

    • If the scale only ranked effectiveness levels (e.g., Poor, Average, Excellent), it would be ordinal.

  • A true zero in a ratio scale means the complete absence of what is being measured.

  • Sleep is a biological function that fluctuates and cannot be completely absent forever, it does not qualify as ratio data. Therefore a report of "0 hours of sleep," does not mean they permanently lack the ability to sleep or that sleep is an absolute quantity that can be "absent" in the same way as weight or money.

Parameter and Statistic

Aspect

PARAMETER(\mu)

STATISTIC(\bar{X})

Definition

A numerical value that describes a population characteristic.

A numerical value that describes a sample characteristic.

Scope

Refers to the entire population.

Refers to a subset (sample) of the population.

Purpose

Represents the true value of a population characteristic.

Used to estimate the population parameter.

Example

The average IQ of all university students in the country (\mu).

The average IQ of 200 randomly selected university students (\bar{X}).

Descriptive vs. Inferential Statistics

DESCRIPTIVE STATISTICS

INFERENTIAL STATISTICS

Description

Summarizes and describes the characteristics of a dataset.

Makes predictions or generalizations about a population based on sample data.

Data Used

Uses sample or population data.

Uses sample data to estimate population characteristics.

Measures

Measures of Central Tendency (Mean, Median, Mode), Measures of Variability (Range, Variance, Standard Deviation), Data Visualization (Graphs, Tables, Charts).

Estimation (Confidence Intervals), Hypothesis Testing (t-tests, ANOVA, Chi-square tests), Correlation and Regression Analysis.

Scope

Can describe a full dataset if the entire population is measured.

Uses a sample to infer about a larger population.

Sampling Error

No concern for sampling error (as it describes data at hand).

Subject to sampling error, as conclusions are drawn from a subset of the population.

Sampling error

Sampling error is the naturally occurring discrepancy, or error, that exists between a sample statistic and the corresponding population parameter.

  • Example of what could go wrong:

    • A university wants to measure overall student satisfaction but only surveys 200 students out of 10,000.

    • Potential biases include overrepresentation of honors students, underrepresentation of freshmen and transfer students, most respondents being from one department, and the survey being conducted only in the library.

Sampling Procedures

Probability Sampling
  • Random Selection: Used to ensure every individual has an equal chance of being selected, increasing generalizability.

METHOD

DESCRIPTION

PSYCHOLOGY EXAMPLE

Simple Random Sampling

Random selection where every person has an equal chance of being chosen.

A researcher randomly selects 200 students from the university’s enrollment list to study the relationship between sleep quality and academic performance.

Systematic Sampling

Selecting every k-th person from a list.

A study on the effects of social media on student focus selects every 10th student from the university's roster.

Stratified Sampling

Dividing the population into subgroups (strata) and sampling proportionally from each group.

A study on test anxiety ensures equal representation of freshmen, sophomores, juniors, and seniors in a university.

Cluster Sampling

Selecting entire groups (clusters) instead of individuals.

A researcher studying academic burnout selects entire psychology departments from different universities.

Multistage Sampling

Combining multiple probability sampling techniques in stages.

A study on mental health in rural communities first randomly selects provinces, then towns within those provinces, then individuals within those towns.

NONProbability Sampling
  • NON Random Selection: Not all individuals have an equal chance of being selected. Often used when random selection is not feasible.

METHOD

DESCRIPTION

PSYCHOLOGY EXAMPLE

Convenience Sampling

Selecting individuals who are easy to access.

A professor surveys their own psychology students about study habits.

Purposive (Judgmental) Sampling

Choosing participants based on specific characteristics.

A study on schizophrenia only includes diagnosed patients from psychiatric clinics.

Snowball Sampling

Using participants to recruit others in hard-to-reach populations.

A researcher studying survivors of domestic abuse asks participants to refer others who have similar experiences.

Quota Sampling

Ensuring specific proportions of groups, but selection is non-random.

A study on gender differences in workplace stress ensures 50% male and 50% female participants but recruits them through social media rather than random selection.

When to Use Random vs. Non-Random Sampling

USE RANDOM SAMPLING

USE NON-RANDOM SAMPLING

Purpose

To generalize findings to a larger population.

To explore specific groups or experiences.

Bias

Eliminates bias by giving everyone an equal chance.

May have selection bias, but useful for targeted research.

Population Characteristics

Large, well-defined populations.

Small or hard-to-reach populations.

Requirements

Requires more planning, data collection tools.

More flexible and cost-effective.

Research Type

Quantitative research, surveys, experiments.

Qualitative research, case studies, exploratory studies.