✅🌸 Unit Zero - Research Methods

Define and give an example of a hypothesis

statement that serves as a tentative explanation; always presented so that it is FALSIFIABLE (can be supported or rejected)

ex: being around trusted friends and family members leads to increased levels of happiness (can be tested - an argument)

What is meant by falsifiable

term meaning that a hypothesis: can be supported or rejected

- able to be proven false

Define operational definition and give an example

a precise description of your variable(s); and how you will numerically measure it

ex: SAT score for intelligence

Define and give an example of each quantitative and qualitative data

quantitative: numerical measures + LIKERT SCALEEEE

> ex: how many times a person performs acts of service for helping behavior

qualitative: info that cannot be counted/measured/numerically represented

> ex: peoples reaction to a certain food

Why are operational definitions necessary (3 reasons)

1. allow multiple researchers working on a study to collect accurate and reliable (CONSISTENT) data

2. allows a study to be replicated: exactly repeated in the future

3. helps the peer review process

What is descriptive research? What does it include? What is its purpose?

to describe and gather info based on observing behaviors

includes: naturalistic observation + case studies

purpose: observe and record behaviors

Describe and give an example of and list the advantages and disadvantages: naturalistic observation

observe behaviors in their natural environment WITHOUT interaction

ex: monkeys in natural environment OR how kids play on a playground to study aggressive behaviors

---

pros: looks at situations that cannot be done in a lab (ppl would change their behavior if so)

cons: doesn't reveal cause/effect (why a behavior happens) + might be UNETHICAL to conduct

Describe and give an example of and list the advantages and disadvantages: case studies

a single case (1 person like Kitty, or a group of people who are the same and extremely unique like a cult)

STUDIES IN DETAIL

pros: can obtain vast amount of info about a unique situation (allows you to study things that are impossible to study)

cons: self-report bias from SURVEYS AND INTERVIEWS due to social desirability + wording effects

ALSO cannot generalize the result to the rest of the world -- just that one specific case

What is self-report bias?

(errors due to) asking people to describe their thought, feelings, or behaviors INSTEAD of measuring these directly and objectively

> can result in incorrect answers due to social desirability or wording effect

What is meant by social desirability?

people lie bc they want to look "good"

What are wording effects?

the way a question is phrased -- can alter the answers you get

What is a correlation?

the relationship between two variables

What letter is used to represent correlation?

r

What is the range of correlation coefficients

-1.00 to +1.00

Draw 3 scatterplots and assign a numerical value to: negative correlation, positive correlation, no correlation

pos: r = 0.99 (factors move in same direction)

neg: r = -0.62 (factors move in opposite direction

none: r = 0

How does the clustering of the dots on a scatterplot indicate strength

if the dots are closer together, it means that the 2 variables have a stronger correlation/relationship

if they are farther apart, they have a weaker relationship

What indicates the strength of the relationship? What indicates the direction of the relationship?

1. the NUMBER VALUE indicates STRENGTH of the relationship

2. the + or - indicates the DIRECTION of the correlation between the two relationships

Miles has discovered that his dog pees on the floor less often when he his home. What kind of correlation is this and why? Bob tries less hard on his homework when he has less motivation. What kind of correlation is this and why?

1. Negative correlation because he is more at home, the dog pees less

2. Positive correlation because variables will both increase in the same direction together

How are correlational studies done

1. take 2 existing data sets

2. collect QUANTITATIVE survey data

then compare the data sets to see how well they predict each other

Why do we use correlational data?

1. to predict outcomes (ex: GPA predicts college successes)

2. look at data that cannot be ethically manipulated in a lab (Ex; depression)

3. can lead to further experimentation (ex: smoking to lung cancer)

What is meant by correlation does not equal causation?

because we don't control the variables of the relationship, there could be another thing affecting the relationship or the relationship could be by chance

THEREFORE we cannot say its a cause and effect relationship -- only predication

What is the directionality problem?

we don't know which variable is responsible for the relationship

ex: low self esteem could cause depression or vice versa

Define and give an example of third-variable problem

a 3rd intervening variable that causes the relationship between two variables

ex: distressing events/biological predisposition COULD cause the relationship between low self esteem and depression

What is an experiment and why is it best?

a scientific procedure to determine something or test a hypothesis

in experiments: we purposefully manipulate a condition in order to determine a CAUSE and EFFECT relationship -- correlations only show possibility

Define and give an example of: independent variable

variable that is purposefully manipulated/changed (typically in groups)

ex: studying with a type of music -- rap music, country music, pop music, no music

Define and give an example of: dependent variable

variable that is being measured/directly influenced by the independent variable

ALWAYS NUMERICAL

ex: the focus level during a study session

Define and give an example of: experimental group

the group(s) that receive treatment/is manipulated in some way

ex: the rap, country, pop music groups

Define and give an example of: control group

the group that does not receive treatment or is not manipulated/changed in any way

ex: the no music group

Define and give an example of a confederate. When are confederates used?

someone who is acting as another participant in the study (a fake or STOOGE)

ex: 2 calm people n a smoke situation to act as a participant

- used when the researchers need to see what the real participants will do in complex SOCIAL settings

> will help/report back to the researcher

How many experimental and control groups can/should you have?

experimental: several but ideally no more than 4

control group: ONLY 1

Why is the control group treated identical to the experimental group minus the treatment

to give an accurate representative of what they are testing and see the difference between the experimental group

What is a confounding variable? Give an example

also known as confound

-- anything that could alter your data (known or unknown)

> gives erroneous/wrong results in your analysis

ex: age (in medicine, etc), amount of sleep

Define and give an example of: experimenter bias and participant bias

occurs when the experimenter (consciously or not) has certain expectations about an outcome -- alters their behavior and impacts the results

ex: a researcher having participants act/respond in a certain way

----

occurs when the participant (consciously or not) has certain expectations about an outcome -- alters their behavior and impacts the results

ex: someone taking a drug that makes them sick, then experience as a result

Define and give an example of: double blind experiment

neither the participants nor the researcher are aware of which condition participants are in (i.e. experimental or control)

ex: 5 people take a placebo, 5 other people take a pill to help with insomnia (but don't know if they took the "real" pill, neither does the researcher)

What is a placebo? What is the placebo effect

a drug w no active/real ingredients

can cause the placebo effect: any observed effect on behavior/health that is caused by a placebo

Define and give an example of: single blind design

when the participants don't know which condition they are in but the searcher does

Ex: researcher knows which medicine is real and which is the placebo but the participants don't

Why is double blind preferred to single blind

to eliminate experimenter + participant bias

Define and give an example of: random assignment

assigning participants to the experimental/control group through randomization procedure

each participant has an EQUAL chance being placed in either control or experimental group

ex: a mix of smart + dumb kids, golfers, etc, in a group & not all smart kids, golfers in their own group

What is random assignment -- HOW do we randomly assign

Why do we use random assignment? Give an example

ensures one group doesn't accidentally introduce an extra independent variable (minimize individual differences) by spreading them out in each group

- typically done with a random # generator

ex: not putting all girls in one group

Name two advantages and two disadvantages to experimental research

pros: allows cause/effect + can control/manipulate outside variables

cons: somethings are unethical to experiment on, may have bias (researcher and participant)

Why are experiments the ONLY design type we can use cause/effect statements with

experiments allow for researchers to control the independent variable, allowing them to see the difference between the results of the treatment

Define and give an example of: population and sample

population: entire group of interest

> ex: all the students from a school

sample: subset of a population

> ex: 50 students of the school

Why do we use the sample and not the population

because including everyone (a population) would take too long

Define and give an example of: random sampling

also known as random selection

each member has an equal chance of inclusion in the study

ex: choosing names of 25 employees out of hat from a company of 250 employees

Define and give an example of: representative sample

a sample that matches your population in the demographics/what you trying to compare to

ex: gamers of all ages

Define and give an example of: convenience sampling

selecting participants based on availability/accessibility to the researcher

ex: participants selected based on their willingness to participate

Why do we need random sampling and representative sample?

we need random sampling so researchers can make conclusions, and generalizations about a specific population without bias

we need representative samples so generalization of a specific population can be made

What is meant by GENERALIZABILITY

the conclusions from a study based on a sample population to the full population

What does random sampling allow you to say? What does random assignment allow you to say?

random sampling allows for generalization for a population

random assignment allows for cause and effect conclusions (groups)

What is the IRB? Why was it established

the Institutional Review Board

- protects safety/welfare/rights of human participants in studies

What are the FIVE (5) Considerations of ethical research on humans

What is meant by informed consent, debriefing and informed assent

1. informed consent (agree to be in study, can leave any time)

> informed assents: minors and their parents consent

2. confidentiality/anonymity

3. debriefing (Explain what happened in the study when done)

4. use of deception must be justified must be followed by a full debriefing)

5. do no harm (physical or mental)

nonmalficence - no harm

What is the IACUC

Institutional Animal Care and Use Committee

(Equivalent to the IRB but for ANIMALS)

- use of animals must be justified and minimized

How do animal ethics differ from human ethics

animal ethics focus on minimizing harm done the animal while human ethics focus on informing them (human) what is happening/happened in the study

What is the purpose of descriptive stats

describe + summarize data

- allows researchers to gain insights + understand patterns, trends, etc

HOWEVER doesn't tell us anything about the population -- only what we saw in the sample

What is the difference between central tendency and variation of the distribution

central tendency: where most of the data is found (mean, median, mode)

- VERTICAL

variation of the distribution: spread and how far the data points lie from each other on the range (range, standard deviation)

- HORIZONTAL

What are the three measure of central tendency -- define each. When do you use each and how do you calculate each

mean: average (most used measure in research) -- used to determine the general value of a data, best for symmetrical data + large data sets - USED IN NORMAL DISTRIBUTION

> add all numbers, then divide by the sum of the number of values in the set

median: # in middle -- used when data is skewed/has outliers, best for asymmetrical data

> order the numbers from least to greatest, the number in the middle is the median - USED IN SKEWED DISTRIBUTION

( if there are an event amount of numbers, add the middle ones and divide by 2)

mode: # that occurs in the MOST often -- used to determine the most frequent value of the data)

> find the numbers that appear the most often

Describe a normal distribution. Label the mean, median, and mode on it

data is symmetrical distributed - no skew

Describe a positive skew. Label the mean, median, mode

the tail on the right is longer (higher mean values)

(mode highest, median, mean lower as in a graph sense)

What is an outlier and what impact does it have on data

data point that varies significantly from the majority of the data

impacts data by distorting the mean, leading to a lower rep. of the sample + distributing skews w/ distribution

What's the range? What the disadvantage to the range -- why do we not use it

difference between lowest and highest scores in the distribution

- doesn't take account of where most of the data is (central tendency) and is impacted by outliers

How do you find the range

subtract biggest value from lowest value to find range

ex: 1,2,3,4,5 -> 5-1 = 4 (Range)

What is standard deviation -- what does it tell you

the average distance scores are spread from the mean/dispersion in a data set

tells us how spread out our data is

ex: bigger SD = more spread (more variability), low SD = less spread from mean

What is the percentile rank and how do we interpret it

the % of the scores in a distribution that are equal or less than a given score

1st percentile (low, better than 1% of group)

99th percentile (high, better than 99% of the group)

ex: student scores in 75th percentile on a test, higher than 75% of test takers

What are inferential statistics

allows you to take data and generalize to the population

What is meant by statistical significance

means your results from the data is NOT due to chance

What does p < .05 mean

your data is statistically significant n data is not due by chance

What is regression towards the mean

means the bigger the sample size, the closer you get to a true avg/greater the effect size

What is effect size

measures the meaningfulness of a magnitude of a relationship between 2 variables -- "practical significance"

closer to 1 = larger the effect size

Define and give an example of: confirmation bias

tendency to search for info that supports our preconceptions and ignore contradictory evidence

ex: asking ppl if they felt the test was hard then finding someone that says it was hard

Define and give an example of: hindsight bias

cognitive bias - people overestimate their ability to predict outcomes after they have occurred

ex: "i knew it was going to rain cause my bones hurt today"

Define and give an example of: overconfidence

tendency to overestimate our knowledge and abilities in a certain area

ex: when a student doesn't study but thinks he's going to get a 100 on the test

Define and give an example of: Hawthorne effect

people modify their behavior when they know they're being watched

ex: a group of students changing what assignment they're working on bc their teacher came to their table

What impact does sample size have on research

larger sample size = more likely there will be statistical significance

What is peer review necessary

other researchers have to check validity and add/question content within the data to ensure it is a well executed study before being published

What is sampling bias? What's an example

when the chosen sample does not represent general population

ex: research done in only male rats (gender bias)

What are cultural norms? How do they impact research

behaviors + thoughts/characteristics of a particular cultural group

impact research bc a particular individual expresses feelings differently which can affect the study

What is the replication crisis - why does it matter

concerns about credibility and reproducibility of social science

- it matters because earlier studies tend to be inaccurate due to false representation, manipulation of statistics, lack of peer review etc

What is meta-analysis and what's it purpose

combining the results of individual studies with statistical techniques to verify findings

purpose: able to combine several studies with small sample sizes in order to make a bigger sample size to make conclusions w more reliability

Describe a negative skew. Label the mean, median, mode

data where the tail is longer on the left (lower values)

better cause mode is higher for test scores

Extra info

"Significant" - P value

Median: less affected by extreme scores/outliers

Positively skewed distribution: mean is higher than median

"Sample is too small" - if it takes about generalizability, LOOK AT SAMPLE SIZE

Most commonly reported measure of central tendency: mean

Most appropriate in determining house values (central tendency): median

Consistency of bowling scores in a season: standard deviation

RANDOM ASSIGNMENT ELIMINATES CONFOUNDING VARIABLES