AP Psyc. Scientific Methods

Independent Variable

The variable manipulated by the experiment (Cause)

Dependent Variable

The variable being measured (Effect)

Confounding Variable

Factors that could also be responsible for changes in the DV that have nothing to do with the IV

Control Conditions

A control for any confounding variables

Random Assignment

Assignment of subjects to each group must be random to eliminate pre-existing differences between those assigned to the different groups

Experimental Group

The group exposed to the independent variable 

Control Group

The group exposed to the independent variable

Placebo

A substance (usually a sugar pill) that may be administered to the control group

Placebo Effect

When the subject thinks they are getting the real IV, not a placebo, and so their body acts accordingly, as if they are just getting the sugar pill

Single Blind Experiment

When the subject does not know what the hypothesis is or if they are in the control group or experimenting group

Double Blind Experiment

When neither the subject nor the experimenter knows what group the subjects are in (avoids the placebo effect and experimenter bias)

Replication (to get similar results each time)

Goal of experiments

Pros of an Experiment

1) only methods to conclude a cause and effect relationship

2) Uses many controls for bias

Cons of an Experiment

1) Artificial setting might cause subjects to behave differently then they would in a natural setting

2) Hard not to know your not in an experiment even though you might not know the hypothesis 

3) Ethical consideration in creating some “real life” situations

Ethical Consideration

1) Must get consent from the subject and they must know about the nature of the study, any risk associated, and having the right to withdraw

2) Subjects can only be hurt if they agree to it, and if the benefit outweighs the risk

3) No names or personal info can be used 

4) If deception is used they have to be told the true nature of the experiment afterwards

5) Use of Institutional Review Board (IRB)

Empirical Evidence

Data that is the result of objective observation, measurement and experimentation 

Pseudoscience

Sciences that make claims on little to no scientific evidence

Hindsight Bias

The tendency to exaggerate one’s ability to have foreseen how something would turn out after knowing the outcome (I knew it all along phenomenon)

Overconfidence

Tendency to overestimate the accuracy of our current knowledge (weare more confident then we are correct) 

Rule of Falsifiability

In order to scientifically test a claim, there must be identifiable evidence that could prove the claim false

Conformation bias

Our tendency to search for information that confirms our beliefs and ignore those that do not

Researcher Bias

The tendency to notice evidence which supports one particular point of view or hypothesis

Volunteer Bias

People who volunteer to participate in a survey are different from those who do not

Participant Bias

Tendency of research subjects to respond in certain ways because they know they are being observed

Social Desirability Bias

The tendency of subjects to present themselves in a socially desirable light

Barnum Effect

Statements that can gull people into thinking they have been accurately assessed by the speaker or test when in fact the outcome could apply to anyone (not verys specific do it can apply to almost anyone)

Operational Defination

1) a definition of a variable in terms of how the variable will be manipulated, measured or observed

2) Must be defined precise, measurable, and concrete

Correlational Method of Research/ Correlation studies

Statistical technique used to measure the strength and nature of the relationship between 2 variables (ONLY FOR SURVEYS) 

Helps in making prediction

Correlation Coefficient

The numerical indication of magnitude and direction of the relationship between 2 variables

Positive Correlation

two variables vary systematically in the same direction

• perfect positive correlation = +1.0

Negative correlation

Two variables vary systematically in opposite directions

• A perfect negative correlation =-1.0

Zero Correlation

There is no relationship whatsoever between 2 variables

Illusory Correlation

When we receive a relationship between variables when none exists (Ex. Believing people in the city are loud after meeting one person from the city) 

Descriptive Research Strategy 

Strategies for observing or describing behavior

Experiment Research Strategy

Only way to try and prove cause and effect

Case Study

Use one or a few individuals to study in depth 

Pros- Good when you have something new, rare, or unusual to study 

Cons- Cannot generalize to population because only one person was studied 

Naturalistic Observation

Observe subjects in their natural environment

Pros- Ethical, see subjects in natural environment

Cons- If they know they are being watched, they may act differently and might not see everything

Longitudinal Study

Study the same person over many years

Pros- Better control over subjects because you are using the same people

Cons- Time-consuming, expensive, subjects might leave or drop out

Cross-sectional Study 

Study subjects if different ages all at the same time

Pros- inexpensive, quicker to perform 

Cons- cannot account for all the different individual variables 

Ex Post Facto

Subjects are selected based on a condition or variable they already have. Usually unethical to manipulate that variable naturally

Pros- Allows more topics to be studied

Cons- No as random (Could be biased)

Meta-Analysis

A procedure for statistically combining the results of many different research studies

Surveys

Research strategies that relies on self-reports, will often use questionnaires, interviews

Pros-  Quick, cheap, can gather large amount of data, confidential 

Cons- No always filled out, not always done honestly, some can have vague or unclear terms, some can lead the subject to answer in certain ways 

How to make a good survey

1) Make it short (15 questions mx)

2) Always include sponsorship (how and why your doing this survey)

3) make it easy to read (clear)

4) Clearly state confidentially

5) Include a follow up procedure

Population

All the cases (people) in a group from which the experiment subjects may be draw and to whom the results apply (ex westford residents) 

Sample

The actual participants in your study (A subgroup of your population) 

Representative sample

Sample needs to be selected so that they represent all traits in the population

Sample Bias

When your sample is not representative of the population and not everyone has an equal chance of getting picked, there is said to be ______

Random Sampling 

every member of the population and not everyone has an equal chance of getting picked

Simple Random Sampling

Randomly selecting individuals name

Cluster Random Sampling

Randomly select naturally occurring groups (Classes, High Schools, Towns)

Stratified Random Sampling

Used to ensure that different groups in a population are adequately represented in a sample 

• Ex. First divide the population into groups, and then randomly select people from those groups

  • Gender, ethnic groups, etc

Non Random/ Convenience Sampling

• When everyone in the population did not get an equal chance to be in a study

• Selecting whoever is available or happens to be in the area

  • Leads to sample bias

Polygon

Plot a point at the intersection of each score (horizontal axis) and its frequency (vertical axis) and then connect the dots with a straight line

Symmetrical Distribution 

Polygon where scores fall equally on both halves of the graph 

Skewed Distribution

Polygon that is asymmetrical in shape scores fall to one side

Positive Skew

Most people had low scores - tail is in the high #s

Negative Skew 

Most people had a high scores- Tail is the low #s

Histogram

Bar Chart with the same horizontal and vertical labels where shaded bar reach up to the frequency score and always touch

Mode

most common scoreM

Mean 

Arithmetic average

Median

Middle Number

Measures of variability

Gives us a single number that presents us with information about how spread out the scores are

Range 

Difference between highest and lowest score( very susceptible to extreme scores) 

Standard Deviation

How spread out is the data?

• the larger the number the more spread out

  • the smaller the number the consistent the scores are to the mean

Z-scores

A number expressed in standard deviation units that shows an individual score’s deviation from the mean. (Shows how you did compared to everyone else)

• +Z score=you are above the mean

• -Z score=you are below the mean

P-Value

The likelihood a result is caused by chance (are the results significant)

• P=0.05 (5%)

Effect Size 

Measures the strength of the relationship between two variables on a numerical scale

Generalizability

The extent to which results of a study can be applied to the outside world.  

Demand Characteristics

 Aspects of the study that suggest to the subjects what type of behavior is expected or desired by the researchers.

Percentile Rank

 Reflects the percentage of subjects who score lower than the subject in question