psych unit 0 ethics and stats

Quantitative Research

Definition: Research that collects numerical data that can be measured and analyzed statistically.

Historical significance: The use of this method helped psychology become a more scientific discipline by allowing researchers to objectively measure behavior and compare results.

Example: A psychologist measures participants' reaction times in milliseconds after exposing them to different stimuli.

Qualitative Research

Definition: Research that collects non-numerical information, such as descriptions, interviews, observations, or personal experiences.

Historical significance: These methods became particularly important for understanding complex human experiences that cannot easily be reduced to numbers.

Example: Interviewing teenagers about how they experience academic stress and analyzing common themes in their responses.

Likert Scales

Definition: A rating scale used in surveys to measure attitudes, opinions, or levels of agreement, commonly ranging from something like 1 = strongly disagree to 5 = strongly agree.

Example:

"I feel confident before taking an exam."
1 — Strongly disagree
2 — Disagree
3 — Neutral
4 — Agree
5 — Strongly agree

Historical significance: developed by psychologist Rensis Likert in the 1930s, became a major tool for measuring attitudes in psychological research.

Institutional Review

Definition: The process of examining a proposed research study to ensure that it follows ethical standards and protects participants.

Historical significance: Research ethics became increasingly important after unethical human research, particularly during the 20th century, demonstrated the need for formal protections.

Tip: It’s essentially asking: "Is this study ethical and safe enough to conduct?"

Informed Consent

Definition: Participants must be given enough information about a study's procedures, risks, and general purpose to voluntarily agree to participate.

Historical significance: This became a fundamental principle of ethical research following serious abuses in human experimentation.

Example: Before participating, a person is told what they will be asked to do and that they can choose whether to participate.

Important: Researchers sometimes cannot reveal the study's exact hypothesis because doing so could influence participants' behavior.

Protect from Harm

Definition: Researchers have an ethical responsibility to minimize physical and psychological risks to participants.

Historical significance: This principle developed in response to unethical research that exposed participants to unnecessary physical or psychological harm.

Example: A researcher cannot intentionally expose participants to dangerous conditions simply to see how they react.

Tip: Research must have a reasonable risk-benefit balance.

Confidentiality

Definition: Researchers must protect participants' private information and identities from being revealed.

Example: Instead of publishing "John Smith scored 82%," a researcher might identify the participant as Participant 17.

Historical significance: This is essential for maintaining trust between researchers and participants and encouraging honest participation.

Research Confederates

Definition: People who appear to be ordinary participants but are actually working with the researcher and are part of the experimental setup.

Historical significance: These people have been especially important in social psychology experiments because researchers sometimes need participants to believe that a social situation is real.

Important experiment: Asch's conformity experiments used these people to intentionally gave incorrect answers to see whether the real participant would conform to the group.
Tip: They are often used when deception is necessary.

Debriefing

Definition: Explaining the true purpose and procedures of a study to participants after their participation, especially if deception was used.

Historical significance: This became an important ethical safeguard as psychologists increasingly used deception in social and behavioral experiments.

Example: After a conformity experiment, researchers explain that the other "participants" were actually confederates.

Institutional Review Board (IRB)

Definition: A committee that reviews research involving human participants to determine whether the study meets ethical and safety standards.

Historical significance: These became formalized in the United States in response to unethical research involving human participants.
Tip: ___ = ethical approval before research begins.

Descriptive Statistics

Definition: Numerical methods used to organize, summarize, and describe data collected from a study.

Examples include:

• Mean

• Median

• Mode

• Range

• Standard deviation

• Percentiles

Historical significance: These allow psychologists to turn large amounts of data into information that is easier to understand.
___ = describe what the data looks like.

Histogram

Definition: A graph that displays the distribution of numerical data using adjacent bars.

Unlike a typical bar graph, the bars in this graph touch because the numerical intervals are continuous.

Example: A psychologist could create this type of graph to show the distribution of IQ scores among 1,000 participants.

Tip: These graphs help you identify whether a distribution is normal, skewed, or bimodal.

Scatterplot

Definition: A graph that displays the relationship between two quantitative variables using individual points.

Example: Researchers could plot:

• X-axis = hours of sleep

• Y-axis = test score

If students who sleep more generally score higher, this graph would show a positive correlation.

Tip: these graphs are used to examine correlation, not necessarily causation.

Measure of Central Tendency

Definition: A statistic that describes the center or typical value of a data set.

The three major measures are:

Mode + Mean + Median

Mode

Definition: The most frequently occurring score in a data set.

Example:

2, 3, 3, 4, 5

___ = 3

Tip: This is especially useful with categorical data because it identifies the most common category.

Mean

Definition: The arithmetic average of a set of scores.

___=sum of scores / number of scores

Example:

2, 4, 6

___= 4

Important: This is strongly affected by extreme scores (outliers).

Median

Definition: The middle score when data are arranged from lowest to highest.

Example:

1, 3, 5, 7, 9

___ = 5

Tip: This is less affected by extreme scores than the mean, making it particularly useful for skewed distributions.

Percentile Rank

Definition: A person's position relative to others in a distribution, expressed as the percentage of scores below that person's score.

Example:
A score at the 90th percentile means the person scored higher than approximately 90% of the people in the comparison group.

Important: 90th percentile does not mean the person got 90% of the questions correct.

Skewed Distribution

Definition: A distribution in which scores are not evenly distributed around the center, creating a longer tail on one side.

Positive/right skew: Tail extends toward higher scores.

Negative/left skew: Tail extends toward lower scores.

Right/positive skew: Mean > Median > Mode

Left/negative skew: Mean < Median < Mode

Example: Income distributions are often positively skewed because a relatively small number of people have extremely high incomes.

Bimodal Distribution

Definition: A distribution with two distinct peaks or modes.

Example: If researchers combine test scores from two very different populations, the graph might have two peaks.

Tip: This type of distribution may suggest that the data contain two different groups.

Measures of Variation

Definition: Statistics that describe how spread out the scores are within a distribution.

Basic measures : standard deviation, and range

Range

Definition: The difference between the highest and lowest scores.

___ =Highest score−Lowest score

Example: 10, 15, 20, 30

___= 30 − 10 = 20

Tip: This is easy to calculate but can be heavily influenced by extreme scores

Standard Deviation

Definition: A measure of how much scores typically differ from the mean.

• Small → scores are clustered close to the mean.

• Large→ scores are spread farther from the mean.

Historical significance: Standard deviation became one of the most important statistical tools for describing variation in psychological data.

Normal Curve

Definition: A symmetrical, bell-shaped distribution in which most scores cluster around the mean and fewer scores occur toward the extremes.

Characteristics:

• Mean = Median = Mode

• Symmetrical

• Most scores near the center

• Few extreme scores

Example: IQ scores are designed to approximately follow a normal distribution.

Tip: This is extremely important for understanding standard deviations, z-scores, and percentiles.

Inferential Statistics

Definition: Statistical methods used to determine whether results from a sample can be generalized to a larger population and whether observed differences are likely meaningful.

Example: Researchers study 500 students and use their results to make conclusions about a larger population of students.
Tip: ___= What can we conclude from what happened?

Meta-Analysis

Definition: A statistical method that combines the results of multiple studies investigating the same question.

Example: Instead of relying on one study about whether sleep affects memory, researchers combine results from dozens of studies to determine the overall effect.

Historical significance: This allows psychologists to identify patterns across many studies and increases the amount of evidence available for a conclusion.

Tip: Particularly useful for determining whether findings replicate across different studies.

Statistical Significance

Definition: A result is statistically significant when the observed result would be unlikely to occur by random chance alone, according to a predetermined statistical criterion.

A common cutoff is:

p < .05

Important:This result does not automatically mean the effect is large or practically important.

P-Value

Definition: The probability of obtaining results at least as extreme as those observed, assuming the null hypothesis is true.
Tip: This does not tell you the probability that your hypothesis is true.

Remember:

Small one → stronger evidence against the null hypothesis.

Common threshold:

p < .05 → statistically significant

Effect Size

Definition: A measure of how large or meaningful an observed difference or relationship is.

This is different from statistical significance.

Example:

Study A:

• p < .001

• Very small effect

Study B:

• p = .04

• Large effect

Study A has stronger statistical evidence, but Study B could have a more practically meaningful effect.

Reliability

Definition: The consistency and dependability of a measurement or research method.

If a psychological test is reliable, it should produce similar results under consistent conditions.

Example: If someone takes a personality test twice under similar circumstances and gets very similar scores, the test demonstrates good reliability.