AP Psych notes from 4.3 and onwards

What are experiments used for and what is necessary for this to occur?

• experiments are used to establish a cause and effect

• experiment - a research method in which an investigator manipulates one or more factors (iv) and observe the effect or the dv

• they also aim to control other variables/factors by random assignment of participants

null hypothesis - has no effect on exam scores. is hat the effect being studied does not exist. The null hypothesis can also be described as the hypothesis in which no relationship exists between two sets of data or variables being analyzed.

reliability - results are consistne

validaity- results measure what they are supposed to measure

What do experiments enable researchers to do? think factors

Experiments enable researchers to isolate the effects of one or more factors by

1. manipulating the factors of interest

2. holding constant “controlling” when it comes to other factors

• expeirmental group - group exposed to treatment

• randomly assign people to each condition

• if groups differ when the experiment ends, we can infer that treatment had an effect

What is the diffference between random sample and random assignment?

random sampling - creates a representative survey sample

random assignment - equalizes the experimental and control groups

• correlational studies, which uncover natrually occuring relationships, are complemented by experiments, which manipulate the factor to determine its effect

• to determine treatments effect, other factors must be controled, which is exaclty how research studies are done

• one group recieves the placebo effect pill and the other group recieves treatment

single blind procedure- an experimental proceudre in whihc the research participants are ignorant about whether the treatment they received is a plaebo

double bilnd procedure - neither the participants nor those who adminsiter the drug and collect the data will know which group is recieving treatment

the more expensive the placebo, the more real it feels to be

Independent vs Dependebt variables

what are independent, depeendent, and cofounding variables?

• independent variable - the factor that is changed or manipulated, the variable whose effect is being studied

• dependent variable - the outcome that is measured, the varibale that may change when the independent variable is manipulated

• confounding variables - other factors that can potentially influence the studies results, the factor other than the factor being studied that might influence the studies results

• both variables given operational definitions

what do single blind procedures and double blind procedures help reduce? what is experimenter bias? what does random assignment control for?

• single blind - help control for social desirability, which is partiipants affecting the results by trying to please the researcher

• double blind - reduce epxerimenter bias, which is when researches may unintentionally infleunce results to confirm their own beliefs

• in experiments, random assignment ensures that confounding variables have an equal chance of appearing in the experimental and control conditions

• random assigmnet controls for possible confounding variables

to establish causation, experimenters control for confounding variables by randomly assigning some participants to an experimental group and others to a control group

they measure the dependent varibale to determine the effect of the independent variable

validity - it tests what it is supposed to test. example: did the landlord’s responses actually vary with the ethnicity of the name?

reliability - the extent to which findings can be replicated

recap:

• variable is anything that can vary

• expeirments manipulate an independent variable, measure a dependnet variable, and control for confounding effects

• an experiment has two conditions, which are experimental conditions or comparisson or control condition

• random assignment works to minimize preexisting diffeences between the groups before any treatment effects occur

• random assignment - helps minimzie preexisting differneces between two groups

0.5 - research design and ethics in psychology

scientists consider the most apporpirate research design, for example, it would not be ethical to place some children into loving homes and others into not loving homes

• quantitative research - methods use numerical data to represent degrees of a varibale, for example using a likert scale, or questionare responses that fall on a continuum like strongly agree ro disagree

• a research method that relies on quantifiable, numerical data

• qualitiative research - rely on relative narrative data, like strcutured interviewsto understand the causes and consequences on the behavrio of individuals

• relies on in depth, narrative data that are not translated into numbers

• an experiments purpose is not to re-create the exact behaviros of every day life but to test theoreticla thoeries

• its the resulting principles, not every day findings that help explain every day behaviors

• many principesl derived in the l abraotry generalize to everyday world life

• researchers sometimes decieve people or withold information from them, but only if it is ethical

todays ehtic codes require:

1. obtain potentila participants informed consent

2. to protext particpats from greaten-than-usual harm and discomfort

3. keep inormation about paprticipaints confidential

4. fully debrief people (psotexperimental explanation of a study, including its purpose and any deceptions to its participants)

curioity, perserverence, nd honesty are most important

how do psychologists’ values infleunce what they study and how they apply their results?

• influence choice of topics

• values can “color” the facts and we sometimes see what we expected to see

• values infrom psychological scinece and scinece has the power to persuade

0.6 - statisitical reasoning in every day life

• critical thinking must be used when presented with big, round, undocumented ideas and numbers

what are descriptive statistics?

• descriptive statistics are numerical data used to describe characteristics of groups, include measures of central tendency, and measure variation ex: techerrs using statistics to see how students have performed

• after organizing and describing data, researches summarize the data using central tendnecy - a single score that represents a whole set of scores

• mode is the simplest measure, is the most frequently occuring socre or scores

• midpointn is median

• half will be above median and half below

• mean and median tell different true stroies, always consider what central tendnecy is reported

• it if its mean, consider whether a few atypiccal socres could be changing it

what is the relative usefulness of the tow measures of variation?

• the range of scores can sometimes proivde a too big estiamte for variation, standard deviation is better to measure

• shows if scores are packed together or disperesedbecause it incorporates ifnormation from each score

• most cases fall next to the mean, less fall toward the extreme

• standard deviation - a computed way to measure how much scores vary around the mean score

• normal curve - a bell shaped symmetrical curve that describes the disitrbution of many types of data, with most scores falling near the mean

• also called normal dsitribution

the dependent variable is always on he y axis, and the independent variable alwasy goes on the x axis

what are infernetial statisitics?

• help us determine if results can be generalized ot a larger population

• numerical data that allow one to generalize from sample data the porbablitity of something being true of a population

0.6-5

how do we determine whether an observed difference cna be generalized to toher populations?

1. representative samples are better than biased unrepresentative samples

2. bigger samples are better than smaller ones- average baed on many cases are more percise than averages based on.a few. larger samples also make replication more likely

3. more estimates are better than fewer estiamttes - a best thing to do is to use meta-analysis (statisitc way of analuzong the results of multiopke studies and reaching an overall conclusion)

estimated based on only a few cases are imprecise

• Essentially, descriptive statistics state facts and proven outcomes from a population, whereas inferential statistics analyze samplings to make predictions about larger populations.

• statsitical testing is used to estimate the probablity of the result occuring by chance

• null hypothesis - assuming that no difference exists beteen groups

• statistically significant - a statistical statement of how likely it is that a result, such as a differnce btween samples, occured by chance, assuming there is no difference between the populations being studied

when the difference we estimate is large, it is also more likely to be more generalizable

“very low” is considered to be p <0.05

• a statsically singifincat result may have little practical significance

• when the result is large they may be statisitical singificance byt a tiny effect size

• effect size - the strength of the relationship between two variables. the larger the effect size, the more one varibale can be explained by the other

• confidence interval - a range of values that likely include the poulation’s true mean value.

statistical singifcnacne indicates the likelihood that the result would have happened by chance if the null hypothesis (no difference) were true, but statisitically significnat is not the same thing as important or strong

descriptive statistics summarize data, while inferential statistics determine whether the data can be applied to a more general population

statistics is the study of the collection, analysis, interpreaion, presentation, and organization of data. it is an efficient way to share data.

descriptive statistics - a numerical data used to measure adn describe charactersitics of groups

the bell curve represents norml distribution

the first step is to gsther data

researches summarize the data using some measure of cetnral tendency. this single score that represents the whole set of scores. tthey are mean, median, and mode. these are neat ways to summarize data.

illusory correlation - when we percieve that a relationship exists

correlation does not equal causation

• To isolate cause and effect, researchers must conduct an experiment: a research method in which an investigator manipulates one or two factors (independent  variables) to observe the effects on some behavior or mental process (dependent variables). 

• lacebo effect: experimental results caused by expectations alone (i.e. the participants acts or claims to feel a certain way because they think they have received a drug that would cause that reaction)

• Extraneous/Confounding Variables: factors that impact the dependent variable that are not the independent variable

• Experimental Bias: researcher only notes aspects of the experiment that support their hypothesis, ignoring anything that could challenge their hypothesis

• Validity: the extent to which a test or experiment measures or predicts what it intends to

• Hawthorne effect: the modification of behavior by study participants in response to their knowledge that they are being observed or singled out for special treatment

• Barnum effect: the phenomenon that occurs when individuals believe that personality descriptions apply specifically to them (more so than to other people), despite the fact that the description is actually filled with information that applies to everyone.

corrleations research - used to understand relationship between two varibales, scatter plots display info

• correlation coefficinet measures how strong the relationship is, the closer to one it is the stronger the correlation

• 0 through -1 fi it is a negative relationship