Research Methodology is more exciting than it sounds. It provides the framework for conducting valid and reliable studies.
Ensures findings are credible and trustworthy.
Helps avoid biases and errors in data interpretation.
Scientific Foundations of Psychology (10–14%)
1.2 Research Methods in Psychology
1.3 The Experimental Method
1.4 Selecting a Research Method
1.5 Statistical Analysis in Psychology
1.6 Ethical Guidelines in Psychology
Psychology is a science, and therefore relies on research to validate theories and inform practices.
Psychological interventions should be grounded in empirical evidence.
Research helps identify effective treatments and practices.
Clever Hans was a horse that appeared to do simple math and answer questions, demonstrating the importance of controlled experiments.
Oskar Pfungst, a psychologist, discovered that Hans was picking up on subtle body language from his owner, highlighting the risk of experimenter bias.
When the owner was hidden, Hans could not answer correctly, proving that the horse's abilities were not mathematical but based on unconscious cues.
The goal of psychology is to develop explanations for behavior and mental processes, leading to a better understanding of the human mind.
These explanations are called theories, and are based on empirical studies, ensuring that they are grounded in evidence.
A theory is a testable explanation for a set of facts or observations, and it serves as the foundation for further research and hypotheses.
Testable: Can be evaluated through empirical research.
Falsifiable: Can be proven wrong through observation or experimentation.
Parsimonious: Simple and straightforward.
Comprehensive: Explains a wide range of phenomena.
A 5-step process for empirical investigation of a hypothesis, designed to control biases and subjective judgments, ensuring objectivity.
Based on direct experience and observation, providing a reliable way to gather data.
Three key attitudes:
Curiosity (explore): A drive to investigate and understand phenomena.
Skepticism (double-check): A critical mindset that questions assumptions.
Humility (recognize errors): An acceptance of the possibility of being wrong and a willingness to correct mistakes.
Hypothesis: A statement predicting the outcome of a study or describing the relationship among variables, derived from a theory.
A hypothesis is a "little theory," a specific prediction that can be tested.
A variable is anything that can vary among participants in a study, such as age, gender, or test scores.
Example: Participating in class leads to better grades than not participating, illustrating a relationship between two variables.
Hypotheses must be testable and falsifiable to be considered scientific.
Falsifiability: The possibility that an assertion can be shown false by observation or experiment, a critical component of scientific inquiry.
Allows for the rejection of incorrect theories.
Promotes the development of more accurate explanations.
A theory cannot be considered scientific if it does not admit the possibility of being shown false, highlighting the need for empirical testing.
All good hypotheses need an operational definition to ensure clarity and precision.
An operational definition describes exactly what the variables are and how they are measured within the context of your study, reducing ambiguity.
Explain what you mean in your hypothesis, providing a clear understanding of the variables.
Explain how the variables will be measured in "real life" terms, making the study replicable and verifiable.
How you operationalize the variables will tell us if the study is valid and reliable, ensuring the quality of the research.
Example: Impact of sleep deprivation on driving performance – define sleep deprivation and driving performance, specifying how each is measured.
Example Scenario:
Hypothesis: Chocolate causes violent behavior.
Operational Definitions:
What do you mean by chocolate?
What do you mean by violent behavior?
How can it be measured?
Use precise and measurable terms.
Ensure definitions are clear and unambiguous.
Align definitions with the theoretical concepts being studied.
A hypothesis must undergo rigorous tests before acceptance, ensuring that the results are reliable and valid.
To make a test controlled, account for the independent variable, enabling the isolation of its effects.
Independent Variable:
A stimulus condition that the experimenter changes independently of all other controlled conditions, allowing for the observation of its specific impact.
Whatever is being manipulated in the experiment, making it the potential cause in the cause-and-effect relationship.
Hopefully the independent variable brings about change, demonstrating its influence on the dependent variable.
If there is a drug in an experiment, the drug is usually the independent variable, as it is being tested for its effects.
Allows for the isolation of the independent variable.
Reduces the impact of confounding variables.
Getting information by direct observation that relies only on the independent variable and not on the experimenter’s hopes, ensuring objectivity.
This data is called the dependent variable, as it is the outcome being measured.
Dependent Variable:
The measured outcome of a study, or the response of the subjects in the study, reflecting the impact of the independent variable.
Whatever is being measured in the experiment, providing quantifiable data.
It is dependent on the independent variable, as its value changes in response to the manipulation of the independent variable.
The dependent variable would be the effect of the drug, showing whether the drug had the intended outcome.
Measurable: Can be quantified and analyzed.
Reliable: Consistent across repeated measurements.
Valid: Accurately reflects the variable being studied.
Independent Variable vs. Dependent Variable
Independent Variable (IV) = stimulus or cause
Dependent Variable (DV) = response or effect
Both the IV and the DV must have an operational definition to ensure clear measurement and interpretation.
Explain what each will look like and how it will be measured, providing a comprehensive understanding of the variables.
Consists of looking at the data collected and seeing if it supports or disproves the hypothesis, using statistical methods.
Helps determine the significance of the results.
Can reveal patterns and relationships in the data.
The last step is to have the results withstand the criticism and scrutiny of the science community, ensuring the robustness of the findings.
Critics check each others’ work by replicating the study, sometimes under slightly different circumstances to see if the same results can be duplicated, validating the original findings.
Replicate: To do a study over to see if the same results are obtained. To control for bias, the replication is most often done by someone other than the original researcher, increasing the reliability of the results.
Confirms the validity of the original findings.
Helps identify potential errors or biases.
Applied Research: Has clear, practical applications. YOU CAN USE IT!!!
Basic Research: Explores questions that you may be curious about, but not intended to be immediately used.
Example: Research on therapies for drug addicts (applied).
Example: Studying how kissing changes when you get older (basic).
Developing new educational strategies.
Improving workplace productivity.
Investigating the neural basis of memory.
Studying the effects of meditation on stress.
A kind of research in which the researcher controls and manipulates the conditions including the IV, to determine cause-and-effect relationships.
Steps in designing an experiment:
Hypothesis
Pick Population: Random Selection then Random Assignment.
Operationalize the Variables
Identify Independent and Dependent Variables.
Look for Extraneous/confounding Variables
Type of Experiment: Blind, Double Blind etc..
Gather Data
Analyze Results
Control group: A group that does not receive the experimental treatment.
Random assignment: Assigning participants to experimental and control groups by chance.
Looking to prove causal relationships
Cause = Effect
Laboratory v. Field Experiments
Smoking causes health issues.
Beware of Confounding Variables
Confounding Variables: Variables that have unwanted influence on the outcome of an experiment, making it difficult to determine the true effect of the independent variable.
Other possible explanations for the dependent variable (result), leading to inaccurate conclusions.
Example: If wanting to prove that smoking causes heart issues, lifestyle and family history may also affect the heart, complicating the analysis.
Random assignment.
Matching participants on key characteristics.
Using control variables.
The Challenges of Experiments
Ensuring all groups being tested have the same conditions (control), which can be difficult in real-world settings.
Ensuring subjects are drawn from a population which consists of everyone who fits the description of your test group, requiring careful sampling methods.
Standardized procedures: Keeping all aspects of the experiment the same for all participants.
Random Selection
To ensure we have a group which represents the demographic we want, we must use random selection, minimizing bias.
Random Selection: Each subject of the sample has an equal likelihood of being chosen for the experimental group, or the group which gets the independent variable.
Ex. Names drawn out of a hat.
Benefits of Random Selection
Increases the likelihood that the sample is representative of the population.
Reduces selection bias.
Sampling
To have confidence in results, they need to be taken from a sample of participants chosen in an unbiased manner.
Random Sample: A sample group of subjects selected by chance, or without biased selection techniques.
Types of Sampling
Stratified sampling: Dividing the population into subgroups and randomly sampling from each subgroup.
Cluster sampling: Dividing the population into clusters and randomly selecting clusters to sample.
Random Assignment
Once you have a random sample, randomly assigning them into two groups helps control for confounding variables, ensuring that differences between groups are due to the independent variable.
Experimental Group v. Control Group.
Group Matching: Ensure that experimental and control group are equivalent on some criterions (sex, race, height etc).
Creates equivalent groups at the start of the experiment.
Minimizes the impact of individual differences.
Representative Sample
A sample obtained in such a way that it reflects the distribution of important variables in the larger population in which the researcher are interested- variables such as age, SES, ethnicity, education….
Use a large sample size.
Employ random sampling techniques.
Sometimes we are unable to do experiments for ethical or practical reasons. In this case we must do another kind of research. In this case, we must conduct a different type of research.
Ex post facto: Research in which we choose subjects based on a pre-existing condition.
Ex: Cancer research.
A correlation study is one where researchers try to show the relationship or correlation between two variables (coincidence).
Correlation studies are largely based in statistics, using correlation coefficients to quantify the relationship.
It is important to remember that correlation does not necessarily mean causation, highlighting the need for caution when interpreting results.
If there is no association between two variables, then there is no causal connection, indicating that one variable does not influence the other.
Ex: People who carry lighters are likely to get cancer. (Carrying a lighter will not cause cancer, BUT people who carry lighters are likely to smoke, thus getting cancer).
As more ice cream is eaten, more people are murdered.
Does ice cream cause murder, or murder cause people to eat ice cream?
Illusory Correlation: The perception of a relationship where none exists, leading to false conclusions.
Confounding variables: Other factors that may influence both variables.
Types of Correlation
Positive Correlation: The variables go in the SAME direction. As one increases, the other increases.
Negative Correlation: The variables go in opposite directions. As one increases, the other decreases.
Studying and grades hopefully has a positive correlation. The more you study, the better your grades.
Heroin use and grades probably has a negative correlation. The more you use heroin, the worse your grades.
The closer the coefficient is to +1 or -1, the stronger the correlation.
Method for collecting information or data as reported by individuals, often through questionnaires.
This is a type of data collection known as self-report data, which means that individuals complete the survey (or provide the information) themselves.
Most common type of study in psychology. Large reach and low cost
Measures correlation. Useful for identifying relationships between variables.
Cheap and fast. Can gather data from many people quickly.
Need a good random sample to ensure representative results.
Low-response rate, which can introduce bias.
Keep the survey short and easy to complete.
Offer incentives for participation.
Naturalistic observations are a method where subjects are observed in their natural environment, providing realistic insights into behavior.
Important for subjects to not know they are being observed to avoid reactivity.
Do not manipulate the environment (Pro). Avoids artificiality of lab settings.
Never really show cause and effect (Con), making it difficult to draw causal conclusions.
Privacy concerns: Ensuring that observations do not violate individuals' privacy rights.
A detailed picture of one or a few subjects, providing rich qualitative data.
Tells us a great story but is just descriptive research. Provides in-depth understanding of unique cases
Does not even give us correlation data, limiting generalizability.
The ideal case study is John and Kate. Really interesting, but what does it tell us about families in general?
Limited generalizability: Findings may not apply to other individuals or groups.
These studies are designed to cut down on time and expense, allowing researchers to gather data more efficiently.
In a longitudinal study, one group or subject is studied for an extended period of time to observe changes in the long term. Same subjects for the entire study, providing insights into developmental trends.
Can identify long-term effects and developmental trends.
Cross-sectional studies look at a cross section of the population and studies them at one point in time. Cross-sectional studies examine different groups at one point in time, offering a snapshot of the population.
Ex: No child left behind
Can be conducted quickly and efficiently.
Personal/ Experimenter Bias: When the researcher allows his or her personal beliefs affect the outcome of the study, compromising objectivity.
Expectancy Bias: When the researcher allows his or her expectations to affect the outcome of the study, influencing the results.
Use standardized procedures.
Double-Blind Procedure- A double-blind study is one in which neither the participants or the experimenters know who is receiving a particular treatment, reducing the risk of bias.
This procedure is utilized to prevent bias in research results, ensuring more objective outcomes.
Reduces the impact of both participant and experimenter expectations.
Sampling bias:
The sample is not representative of the general population, leading to skewed results.
Selection bias:
Occurs when the participants in the sample are not equally and fairly selected for both the experimental and control groups; this renders any results from the experiment meaningless
Response bias:
When only highly motivated people return a survey. When this occurs, the resulting data is biased toward those with the motivation to answer and submit the survey, and is therefore not representative of the population as a whole.
Belief Bias:
We make illogical conclusions in order to confirm our preexisting beliefs.
Confirmation bias: The tendency to seek out information that confirms existing beliefs.
The tendency to believe, after learning the outcome, that you “knew it all along”, overestimating your ability to predict events.
After the Chris Brown/Rihanna incident….many people said they knew Chris Brown was a violent person!!! Did they really?
Can distort our understanding of past events.
We tend to think we know more than we do, leading to poor decision-making.
82% of U.S. drivers consider themselves to be in the top 30% of their group in terms of safety.
81% of new business owners felt they had an excellent chance of their businesses succeeding. When asked about the success of their peers, the answer was only 39%. (Now that's overconfidence!!!)
Can lead to risky behavior and poor planning.
Tendency for people to accept very general or vague characterizations of themselves and take them to be accurate, also known as the Forer effect.
False Consensus Effect:
Tendency to overestimate the extent to which others share our beliefs and behaviors
Horoscopes
Some people to work harder and perform better when they are participants in an experiment, due to the attention they receive.
Individuals may change their behavior due to the attention they are receiving from researchers rather than because of any manipulation of independent variables.
Whether the lights were brighter or dimmer, production went up in the Hawthorne electric plant.
Researchers should be aware of the potential for the Hawthorne effect to influence their results.
Placebo effect- Sometimes the act of taking a pill produces an effect if the person believes the pill is active. To compensate for this, scientists often give placebos to determine if an effect is due to the "real" drug or from the act of just taking a pill.
Expectations: The stronger the belief that the treatment will work, the more likely it is to have an effect.
Order Effects- The positioning of question or tasks in a survey, test, etc., influences the outcome. This is designed to measure whether the order of the questions makes a difference in the outcome of the survey.
Randomizing the order of questions or tasks.
Recording the results from our studies, using various statistical methods to analyze and interpret the data.
Must use a common language so we all know what we are talking about, allowing for effective communication and collaboration among researchers.
Descriptive statistics: Summarizing and describing the main features of the data.
Just describes set(s) of data, providing a summary of the main features.
You might create a frequency distribution to show how often each score occurs.
Frequency polygons or histograms can be used to visually represent the data.
Measures of Central Tendency
Mean: The average score.
Median: The middle score.
Mode: The most frequent score.
Frequency Distribution
A summary chart which shows how frequently each of the various scores in a set of data occur, providing a clear overview of the data.
Example: Collecting data on how many hours of sleep college students get each night.
Sort the data in ascending order.
After conducting a survey of 30 of your classmates, you are left with the following set of scores:
7, 5, 8, 9, 4, 10, 7, 9, 9, 6, 5, 11, 6, 5, 9, 10, 8, 6, 9, 7, 9, 8, 4, 7, 8, 7, 6, 10, 4, 8
In order to make sense of this information, you need to find a way to organize the data. A frequency distribution is commonly used to categorize information so that it can be interpreted quickly in a visual way.
In our example above, the number of hours each week serves as the categories and the occurrences of each number are then tallied.
Frequency distributions are often displayed in a table format
Provide a clear overview of the data.
Hours of Sleep Each Night | Frequency |
---|---|
4 | 3 |
5 | 3 |
6 | 4 |
7 | 5 |
8 | 5 |
9 | 6 |
10 | 3 |
11 | 1 |
Total | 30 |
Histogram
A bar graph that uses vertical columns to visually show frequencies, which is how many times a score occurs.
Depicts a frequency distribution where the height of the bars indicates the frequency of a group of scores.
Difference between a bar graph & a histogram:
Histogram = no spaces between bars
Bar Graph= Spaces between each bar
Provide a visual representation of the data.
Frequency Polygon
After a frequency histogram is constructed then the scores at the top-middle of each column are connected by a line- this is a frequency polygon.
It can show the general trend of the frequency of scores by showing which frequencies are most common in a data set.
A line graph made by connecting the top center scores of the columns of a frequency distribution.
Can be used to compare multiple distributions on a single graph.
Mean (average): The measure of central tendency most often used to describe a set of data.
To calculate mean, simply add all the scores and divide by the number of scores.
While the mean is easy to calculate, it has a big downside. It can easily be influenced by extreme scores.
Sum all the values in the data set.
Median: A measure of central tendency represented by the score that separates the upper half of the scores in a distribution from the lower half.
The big advantage of this is the median is not effected by extreme scores, providing a more stable measure of central tendency.
Sort the data in ascending order.
Mode: A measure of central tendency which represents the score that occurs most often.
Count the frequency of each score.
Example
The weekly salaries of six employees at McDonalds are 140, $220, $90, $180, $140, $200.Find: (a) the mean (b) the median (c) the mode
Answers
Mean: (90+ 140+ 140+ 180 + 200 + 220)/6 = $161.67
Median: 90,140,140,180,200,220; The two numbers that fall in the middle need to be averaged. (140 + 180)/2 = $160
Mode: 90,140,140,180,200,220; The number that appears the most is 140
Understanding income distributions.
Standard Deviation (SD): A measure of variability that indicates the average distance between the scores and their mean, showing how spread out the data is.
A low standard deviation indicates that the data points tend to be very close to the mean, whereas high standard deviation indicates that the data are spread out over a large range of values.
A smaller SD indicates that the data is clustered tightly around the mean
Normal Distribution
The standard deviation and mean together tell us a lot about the distribution of scores. A data set with a mean of 50 (shown in blue) and a standard deviation of 20.
MEAN=50
SD=20
A standard deviation of 15 accounts for about 68% of responses. A normal distribution is a bell shaped curve.
Symmetric around the mean
A distribution is skewed if one of its tails is longer than the other. This can affect the interpretation of the mean and median.
The first distribution shown has a positive skew. This means that it has a long tail in the positive direction.
The second distribution has a negative skew since it has a long tail in the negative direction.
Finally, the third distribution is symmetric and has no skew (normal distribution).
Are the results positively or negatively skewed?
In positively skewed distributions, the mean is greater than the median
(Example graph showing mode, median, and mean)
Income distributions are often positively skewed
Correlation: A relationship between two variables in which change in one variable are reflected in the changes in the other variable.
Correlation Coefficient:
A number between –1 and +1 expressing the degree of relationship between two variables.
Shows the strength of the relationship
The relationship gets weaker the closer you get to zero.
+1 indicates a perfect positive correlation
A statistical measure of the extent to which two factors relate to one another
Indicates direction of relationship (positive or negative)
Indicates strength of relationship (0.00 to 1.00)
If the correlation coefficient is a positive number, there is a positive correlation (connection) between the variables.
SAT scores and college achievement—among college students, those with higher SAT scores also have higher grades
If the correlation coefficient is a negative number, there is a negative correlation (connection) between variables.
Education and years in jail—people who have more years of education tend to have fewer years in jail
If the correlation coefficient is 0, there is no correlation between variables.
Positive Correlation
Negative Correlation
No Correlation
Correlation does not equal causation
Correlation Coefficient Comparison
Which is a stronger correlation?
-.13 or +.38
-.72 or +.59
-.91 or +.04
Remember :
The relationship gets weaker the closer you get to zero.
Larger sample sizes provide more reliable estimates of correlation
Scatterplot
A graphed cluster of dots, each of which represents the values of two variables.
the slope of the points suggests the direction of the relationship
the amount of scatter suggests the strength of the correlation
little scatter indicates high correlation
Also called a scattergram or scatter diagram
Each point represents a pair of values for two variables
Tells us how far the score is from the mean, measured in standard deviations.
It is a very useful statistic because it allows us to compare two scores coming from two different distributions, providing a standardized measure.
A positive z score means a number above the mean, indicating the score is higher than average.
A negative z score means a number below the mean, indicating the score is lower than average.
Comparing scores from different tests
The purpose is to discover whether the finding can be applied to the larger population from which the sample was collected, allowing researchers to make generalizations.
T-tests, ANOVA or MANOVA. These tests help to determine the statistical significance of the results.
P-value= .05 for statistical significance. 5% likely the results are due to chance.
T-tests: Used to compare the means of two groups
Research Method | Basic Purpose | How Conducted | What Is Manipulated |
---|---|---|---|
Descriptive | To observe and record behavior | Case studies, surveys, and naturalistic observations | Nothing |
Correlational | Computing statistical association, sometimes among survey responses | Nothing | |
Experimental | To explore cause and effect | Manipulating one or more factors and using random assignment to eliminate preexisting differences among subjects | The independent variable(s) |
Research question
Review serves an important role in the protection of the rights and welfare of human research subjects, ensuring studies are ethically sound.
Group of psychologists or other professionals who look over each proposed research study and judge it according to its safety and consideration for the participants in the study
Reviewing research proposals
Single-site study or review at each site for single site or multisite studies
Example: Most research institutions have at least one IRB at the site that review research conducted at the site
(Flowchart depicting the IRB process)
Submission of research proposal
The participants of an experiment are asked for their agreement to participate, ensuring their autonomy and respect.
The decision to participate should be based on informed knowledge of the experiment along with their rights, promoting transparency.
A guardian or family member should also give consent to the study if the participants are
Children under 18 years of age
Adults incompetent of understanding the true nature and aims of the study
Purpose of the research
Deception should be avoided; however slight deception is permissible if:
Participant bias would result from participants knowing the true aims of the study
The research has potential significant contribution
It is unavoidable
The deception does not cause any distress to the participant, including upon being informed of the deception
If deception is involved, informed consent is not obtained
Any deception must be revealed at the earliest opportunity
When the potential benefits of the research outweigh the risks
At the end of the experiment, the participants should be given a debriefing and should be returned to their normal mental states that they entered the experiment with
Participants are informed about the true