Class Notes

Class Notes: Statistics in Psychological

Research
I. Introduction to Statistics in Psychology
Statistics play a central role in psychological research because they allow researchers to describe
data, test hypotheses, and draw conclusions about behavior and mental processes. Psychology
relies heavily on empirical evidence, and statistics provide the tools needed to determine whether
observed patterns are meaningful or due to chance. Without statistics, psychologists would be
unable to evaluate theories, assess treatment effectiveness, or generalize findings beyond a single
study.

In psychology, statistics are used both to summarize data and to make inferences about
populations based on samples. Researchers collect data from a limited number of participants
and then use statistical methods to determine whether their results likely reflect broader trends in
the population. This distinction between describing data and making inferences forms the
foundation of statistical analysis.

II. Variables and Measurement

Independent and Dependent Variables

An independent variable (IV) is the variable that is manipulated or categorized by the

researcher. It represents the presumed cause in an experimental relationship. A dependent
variable (DV) is the variable that is measured and represents the outcome or effect of the
manipulation.

For example, in a study examining the effect of sleep deprivation on memory performance, sleep
deprivation would be the independent variable, and memory test scores would be the dependent
variable.

Levels of Measurement

Variables in psychology are measured at different levels, which determine the types of statistical
analyses that can be used.
 1. Nominal level
o Categories with no inherent order
o Examples: gender, diagnosis, political affiliation
2. Ordinal level
o Categories with a meaningful order but unequal intervals
o Examples: class rank, Likert-scale responses
3. Interval level
o Equal intervals between values but no true zero
o Examples: temperature (Celsius), IQ scores
4. Ratio level
o Equal intervals and a true zero point
o Examples: reaction time, number of errors

Understanding measurement level is critical for selecting appropriate statistical tests.

III. Descriptive Statistics

Descriptive statistics are used to organize, summarize, and describe data. They do not allow
researchers to make conclusions beyond the data collected, but they provide essential
information about patterns within the dataset.

Measures of Central Tendency

Measures of central tendency describe the typical or average score in a dataset.

1. Mean
oArithmetic average
oSensitive to extreme scores (outliers)
2. Median
o Middle score when data are ordered
o Less affected by outliers
3. Mode
o Most frequently occurring value
o Can be used with nominal data

Each measure provides different information, and the choice depends on the nature of the data
and its distribution.

Measures of Variability
Measures of variability describe how spread out the data are.

1. Range
o Difference between highest and lowest values
2. Variance
o Average squared deviation from the mean
3. Standard Deviation
o Square root of variance
o Indicates typical distance from the mean

High variability suggests greater individual differences, while low variability indicates that
scores cluster closely around the mean.

IV. The Normal Distribution

The normal distribution, also known as the bell curve, is a symmetrical distribution in which
most scores cluster around the mean. In a normal distribution:

 The mean, median, and mode are equal

 Approximately 68% of scores fall within one standard deviation
 Approximately 95% fall within two standard deviations

Many psychological variables, such as intelligence and reaction time, approximate a normal
distribution.

Skewed Distributions

 Positive skew: tail extends to the right

 Negative skew: tail extends to the left

Skewness affects which measure of central tendency best represents the data.

V. Correlation
Definition of Correlation

Correlation measures the strength and direction of the relationship between two variables. It
does not imply causation.
The correlation coefficient, r, ranges from –1.00 to +1.00.

 Positive correlation: variables increase together

 Negative correlation: one variable increases as the other decreases
 Zero correlation: no relationship

Interpreting Correlations

 Strength depends on how close r is to –1 or +1

 Direction indicates whether the relationship is positive or negative

A common mistake is assuming that correlation implies causation. Third variables and reverse
causation must always be considered.

VI. Inferential Statistics

Inferential statistics allow researchers to make conclusions about populations based on sample
data. These methods rely on probability theory and assume that samples are representative of the
population.

Sampling and Sampling Error

A sample is a subset of a population. Because samples rarely perfectly represent populations,

sampling error occurs. Inferential statistics estimate the likelihood that results are due to
sampling error rather than true effects.

VII. Hypothesis Testing

Null and Alternative Hypotheses

 Null hypothesis (H₀): no effect, no difference, or no relationship

 Alternative hypothesis (H₁): there is an effect, difference, or relationship

Researchers attempt to reject the null hypothesis using statistical tests.

p-Values and Statistical Significance

The p-value represents the probability of obtaining results as extreme as those observed if the
null hypothesis is true.

 p < .05 is commonly considered statistically significant

 Statistical significance does not imply practical importance

VIII. Type I and Type II Errors

 Type I error: rejecting a true null hypothesis
 Type II error: failing to reject a false null hypothesis

There is a tradeoff between reducing Type I and Type II errors.

IX. Common Statistical Tests

t-Tests

t-tests compare means between groups.

 Independent samples t-test: compares two separate groups

 Paired samples t-test: compares the same participants across conditions

ANOVA (Analysis of Variance)

ANOVA compares means across three or more groups.

 One-way ANOVA: one independent variable

 Factorial ANOVA: multiple independent variables

ANOVA tests whether group differences are greater than expected by chance.

Chi-Square Test

Chi-square tests relationships between categorical variables.

  Compares observed frequencies to expected frequencies
 Commonly used with survey and diagnostic data

X. Regression Analysis
Regression examines how well one or more variables predict another variable.

 Simple regression: one predictor

 Multiple regression: multiple predictors

Regression provides information about prediction strength and direction.

XI. Effect Size

Effect size measures the magnitude of an effect, independent of sample size.

Common effect sizes include:

 Cohen’s d
 r²

Effect sizes are critical for interpreting real-world importance.

XII. Reliability and Validity

Reliability

Reliability refers to consistency of measurement.

 Test–retest reliability
 Interrater reliability
 Internal consistency

Validity

Validity refers to accuracy of measurement.

  Construct validity
 Internal validity
 External validity

A measure can be reliable without being valid, but not valid without being reliable.

XIII. Ethics and Statistics

Ethical research requires honest statistical reporting. Practices such as p-hacking, selective
reporting, and misrepresenting results undermine scientific integrity.

XIV. Conclusion
Statistics are fundamental to psychological science. They allow researchers to describe behavior,
test theories, evaluate treatments, and make informed decisions based on data. Understanding
statistical concepts is essential for interpreting research findings and becoming a critical
consumer of psychological information.

Class Notes: Statistics in Psychological Research

I. Introduction to Statistics in Psychology
Statistics play a central role in psychological research because they allow
researchers to describe data, test hypotheses, and draw conclusions about
behavior and mental processes. Psychology relies heavily on empirical
evidence, and statistics provide the tools needed to determine whether
observed patterns are meaningful or due to chance. Without statistics,
psychologists would be unable to evaluate theories, assess treatment
effectiveness, or generalize findings beyond a single study.
In psychology, statistics are used both to summarize data and to make
inferences about populations based on samples. Researchers collect data
from a limited number of participants and then use statistical methods to
determine whether their results likely reflect broader trends in the
population. This distinction between describing data and making inferences
forms the foundation of statistical analysis.

II. Variables and Measurement

Independent and Dependent Variables
An independent variable (IV) is the variable that is manipulated or
categorized by the researcher. It represents the presumed cause in an
experimental relationship. A dependent variable (DV) is the variable that
is measured and represents the outcome or effect of the manipulation.
For example, in a study examining the effect of sleep deprivation on memory
performance, sleep deprivation would be the independent variable, and
memory test scores would be the dependent variable.

Levels of Measurement
Variables in psychology are measured at different levels, which determine
the types of statistical analyses that can be used.
1. Nominal level
o Categories with no inherent order
o Examples: gender, diagnosis, political affiliation
2. Ordinal level
o Categories with a meaningful order but unequal intervals
o Examples: class rank, Likert-scale responses
3. Interval level
o Equal intervals between values but no true zero
o Examples: temperature (Celsius), IQ scores
4. Ratio level
o Equal intervals and a true zero point
o Examples: reaction time, number of errors
Understanding measurement level is critical for selecting appropriate
statistical tests.

III. Descriptive Statistics

Descriptive statistics are used to organize, summarize, and describe
data. They do not allow researchers to make conclusions beyond the data
collected, but they provide essential information about patterns within the
dataset.

Measures of Central Tendency

Measures of central tendency describe the typical or average score in a
dataset.
1. Mean
o Arithmetic average
o Sensitive to extreme scores (outliers)
2. Median
o Middle score when data are ordered
o Less affected by outliers
3. Mode
o Most frequently occurring value
o Can be used with nominal data
Each measure provides different information, and the choice depends on the
nature of the data and its distribution.

Measures of Variability
Measures of variability describe how spread out the data are.
1. Range
 o Difference between highest and lowest values
2. Variance
o Average squared deviation from the mean
3. Standard Deviation
o Square root of variance
o Indicates typical distance from the mean
High variability suggests greater individual differences, while low variability
indicates that scores cluster closely around the mean.

IV. The Normal Distribution

The normal distribution, also known as the bell curve, is a symmetrical
distribution in which most scores cluster around the mean. In a normal
distribution:
 The mean, median, and mode are equal
 Approximately 68% of scores fall within one standard deviation
 Approximately 95% fall within two standard deviations
Many psychological variables, such as intelligence and reaction time,
approximate a normal distribution.

Skewed Distributions
 Positive skew: tail extends to the right
 Negative skew: tail extends to the left
Skewness affects which measure of central tendency best represents the
data.

V. Correlation
Definition of Correlation
Correlation measures the strength and direction of the relationship
between two variables. It does not imply causation.
The correlation coefficient, r, ranges from –1.00 to +1.00.
 Positive correlation: variables increase together
 Negative correlation: one variable increases as the other decreases
 Zero correlation: no relationship

Interpreting Correlations
 Strength depends on how close r is to –1 or +1
 Direction indicates whether the relationship is positive or negative
A common mistake is assuming that correlation implies causation. Third
variables and reverse causation must always be considered.

VI. Inferential Statistics

Inferential statistics allow researchers to make conclusions about populations
based on sample data. These methods rely on probability theory and assume
that samples are representative of the population.

Sampling and Sampling Error

A sample is a subset of a population. Because samples rarely perfectly
represent populations, sampling error occurs. Inferential statistics estimate
the likelihood that results are due to sampling error rather than true effects.

VII. Hypothesis Testing

Null and Alternative Hypotheses
 Null hypothesis (H₀): no effect, no difference, or no relationship
 Alternative hypothesis (H₁): there is an effect, difference, or
relationship
Researchers attempt to reject the null hypothesis using statistical tests.

p-Values and Statistical Significance

The p-value represents the probability of obtaining results as extreme as
those observed if the null hypothesis is true.
 p < .05 is commonly considered statistically significant
 Statistical significance does not imply practical importance

VIII. Type I and Type II Errors

 Type I error: rejecting a true null hypothesis
 Type II error: failing to reject a false null hypothesis
There is a tradeoff between reducing Type I and Type II errors.

IX. Common Statistical Tests

t-Tests
t-tests compare means between groups.
 Independent samples t-test: compares two separate groups
 Paired samples t-test: compares the same participants across
conditions

ANOVA (Analysis of Variance)

ANOVA compares means across three or more groups.
 One-way ANOVA: one independent variable
 Factorial ANOVA: multiple independent variables
ANOVA tests whether group differences are greater than expected by
chance.

Chi-Square Test
Chi-square tests relationships between categorical variables.
 Compares observed frequencies to expected frequencies
 Commonly used with survey and diagnostic data

X. Regression Analysis
Regression examines how well one or more variables predict another
variable.
 Simple regression: one predictor
 Multiple regression: multiple predictors
Regression provides information about prediction strength and direction.

XI. Effect Size

Effect size measures the magnitude of an effect, independent of sample
size.
Common effect sizes include:
 Cohen’s d
 r²
Effect sizes are critical for interpreting real-world importance.

XII. Reliability and Validity

Reliability
Reliability refers to consistency of measurement.
 Test–retest reliability
 Interrater reliability
 Internal consistency

Validity
Validity refers to accuracy of measurement.
 Construct validity
 Internal validity
 External validity
A measure can be reliable without being valid, but not valid without being
reliable.

XIII. Ethics and Statistics

Ethical research requires honest statistical reporting. Practices such as p-
hacking, selective reporting, and misrepresenting results undermine
scientific integrity.

XIV. Conclusion
Statistics are fundamental to psychological science. They allow researchers
to describe behavior, test theories, evaluate treatments, and make informed
decisions based on data. Understanding statistical concepts is essential for
interpreting research findings and becoming a critical consumer of
psychological information.