Chapter 2 Research Methods Why do we study research methods 1 Understanding psychological investigation experimentation 2 Understanding when and why certain methods are used 3 Understanding strengths and weaknesses of various designs 4 Understanding the necessity of good research design Important Definitions Heuristics short cuts in thinking Necessary due to resource limitations Can lead to errors in judgment Availability Heuristic estimating the likelihood of an occurrence based on how easily it comes to mind or how available it is to our memories Availability reflects the effectiveness of our search strategy Pros high frequency of events high probability Cons Other things can increase availability too such as recency publicity and distinctiveness External Validity the extent to which findings of a study can be generalized to the real world Internal Validity the extent to which cause and effect inferences can be drawn from a study Random Selection random sampling selection such that every person in the population has an equal chance of being chosen to participate Aims to wash out confounding variables Reliability consistency of a measurement Ex Test and retest reliability ability for a given method or instrument to produce similar measurements at different times Ex Interrater reliability ability for a given method or instrument to produce similar measurements for different individuals Validity degree to which a method or instrument measures what it claims to measure Reliability is necessary for validity because we need to measure something consistently before we can measure it well Types of Studies 1 Naturalistic Observation watching participants behavior in real world settings without trying to manipulate their actions a Pros b Cons i High in external validity findings are relevant to the real world because they are studied in real world scenarios ii No interference so you can study true behavior i Low in internal validity no way to infer cause and effect ii No control over variables so you must wait for them to iii unfold If someone knows they are being observed that can affect their behavior 2 Case Studies examination of one or more person over an extended period of time Utilizes several methods observation experimentation etc a Pros i Provide existence proofs demonstration that a given psychological phenomenon can occur ii Unique opportunity to study rare phenomena that are difficult to recreate in a laboratory iii Can spark new research 2 i Only one source of information so it can be too specific ii Rare phenomena are rarely good iii You have no control over what happens iv You must wait for an opportunity to present itself 3 Surveys and Self Report used to assess characteristics such as personality traits mental illness interests and opinions b Cons a Pros i Easy to administer ii Gets information from individuals directly because observations cannot tell you how a person is actually feeling inside only that person can tell you that iii Gets a lot of data quickly b Cons i They assume that participants have enough insight into their own personality characteristics to accurately report themselves ii They assume that participants will answer honestly iii Response sets developing a certain way to answer different types of questions Ex In interviews you want to make yourself look good Ex Malingering tendency to make ourselves appear psychologically disturbed to achieve a personal goal like getting medicine 4 Correlational Designs examining the extent to which two variables are associated Co together Related associated connected a Pros b Cons i Allows us to generate predictions about the future i We can t be sure why these predicted relationships exist 3 c Positive Correlations 0 as the value of one variable changes the other goes in the same direction d Negative Correlations 0 as the value of one variable changes the other goes in the opposite direction 4 e Zero Correlations the variables don t go together at all they do not correlate f Correlation Coefficients range in value from 1 0 to 1 0 i 1 0 perfect negative correlation ii 1 0 perfect positive correlation iii Values lower than 1 0 a less than perfect correlation iv To find out how strong a correlation coefficient is how closely the variables are related look at its absolute value v or indicates the direction of the correlation EX 27 and 27 are both 27 which means both correlation coefficients are equally large in size and equally informative but they re going in opposite directions g CORRELATION DOES NOT IMPLY CAUSATION 5 Experimental Designs research characterized by 1 random assignment of participants to conditions and 2 manipulation of an independent variable a Allows for cause and effect inferences 5 b Random assignment experimenter randomly sorts participants in order to cancel preexisting differences into one of two groups i Experimental group the group that receives the ii Control group the group that doesn t receive the manipulation manipulation c Blind participants are unaware of which group experimental or controlled they are in i Eliminates experimenter bias d Double blind participants and researchers are unaware of which group participants are in i Eliminates experimenter bias ii Experimenter expectancy effect unintentional biasing of the outcome of a study as a result of knowing the condition assignment e Experiments a test trial or procedure an operation for the purpose of discovering something unknown or of testing a principle i Controlled ii Systematic iii Methodical iv Must have independent and dependent variables f Independent Variables manipulated by the experimenter values that are set beforehand systematically varied values called levels g Dependent Variables what experimenter measures dependent on level of independent variable the response variable what you re interested in finding h When we define our independent and dependent variables for the purposes of a study we re providing an operational definition a working definition of what they re measuring 6 Operationalization process of strictly defining variables into independent and dependent i For an experiment to possess adequate internal validity the ability to draw cause and effect conclusions the level of the independent variable must be the ONLY difference between these groups i Any variable that differs between the experimental and control groups other than the independent variable is known as a cofounding variable j Cons i The Placebo Effect
View Full Document
Unlocking...