Psychpt5

Survey research is based on collecting data using questionnaires or interviews.

Survey research Pros- Efficient way to collect data, can gauge changes across time

Survey Research Cons- Cannot infer cause and effect data, rely on subjective self reports, possible issues with representativeness

 Representativeness- the way certain methods of data collection have the ability to represent specific parts of a population, like how when a magazine put out a survey asking which candidate would win, the results said the conservative candidate because those who can afford magazines tend to be more conservative.

  • Nearly 2 million Americans wre surveyed about who they were planning to vote for in the 1936 presidential election
  • Results predicted alandslide win for republican candiate Alf Landon over democratic candidate Franklin Roosvelt, but Roosvelt won.
  • Why? Survey sample was unrepresentative- Names obtained using telephone directories, car registrations, and magazine subscription lists
  • Created a sample biased towards more wealthy americans.

  Sample- a collection of data.

  • These samples should be representative of the population being studied to increase the probability that inferences made about the sample will generalize to the populaiton that is the target for the research
  • In truly random sampling,

  Stratified random sampling: attempts are made to draw samples in a pseudo-random fashion that takes into account various differences in the population

  samples are constructed to have a composition that more closely resembles the populations being studied. If 50% of the population speak a second laguage, ensure 50% of your sample is bilingual. Correlational research

  • WE can examine correlations using various kinds of methods
  • Correlational research: seeks to examine general relationships between variables of interest for particular research questions and involves measuring at least two variables then determining whether they are statistically related and remember that variables in correlational research are only ever measured, never manipulated.

Correlational Research Pros- gauges the strength of relationships present, can be used to make predictions about variables, indentifies real world associations, can be used to study things we couldn't ethically manipualtion in experiments.

Correlational research Cons- While the fact that these kinds of studies take place outside of the laboratory is a sort of advantage, it also mean they're uncontrolled, these studies don't allow inferences about cause and effect to be drawn, they merely demonstrate associations

Bidirectionality- even if there is some causal relationship, it's not certain which direction it goes (which one causes which?)

Correlations can be mathmatically described by a correlation calculator. Values range from -1 to +1 Sign indicates directions (pos or neg) Absolute values indicates strength of the association.

 Positive correlations- as x increases, y increases and if x decreases, y decreases.

Negative correlation- if x increases, y decreases and if x decreases, y increases.

Experiments- examining cause and effect 1 one variable is manipulated (independent variable) 2 another variable is measured to determine what effect