Chapter 19 Flashcards
ancestry approach
the reference lists of all the articles obtained so far checked to see if they cite any potentially relevant manuscripts that the search missed
Cohen’s d
computed by subtracting the mean of the control group from the mean of the experimental group and dividing by the pooled standard deviation of two groups
combining studies
arriving at an overall, bottom-line answer to the focused hypothesis of interest
comparing studies
analyses that seek to document and explain variability among the effect sizes of the group of studies
differential mortality
a higher percentage of people being dropped from one group compared to the others
diffuse tests (tests of heterogeneity)
the goal is to determine whether there is significant variability among a group of effect sizes using a x^2 test
effect size
the strength or magnitude of an association in the population or degree of departure from the null hypothesis
electronic database
i.e. psycINFO or MEDLINE
focused question
group of studies all address the same hypothesis
focused tests
explicitly test hypotheses about the reasons for variability among effect sizes (advocates z test)
fundamental equation of data analysis
significance level = size of effect x size of study
invisible college
the network of researchers working on the topic
key words
words entered when one searches by subject
literature reviews
thorough search for all research conducted on the topic
literature search
identify the relevant electronic database then decide on key-words
meta-analysis
the results of studies are combined and analyzed quantitatively
moderator variable
third variable that affects correlation of two variables
narrative literature review
studies are described and summarized verbally
psychometric properties
construction and validation of measurement instruments
publication bias
the tendency to overstate the magnitude of a given effect because relevant non-significant findings or failures to replicate are not included in the review
publication lag
the interval between manuscript acceptance and publication // usually exceeds a year
standardized effect-size index
puts the results of all the studies on the same scales in order to compare results across studies meaningfully
statistical power
the likelihood of finding an effect that exists in the population (opposite of type II error)
thesaurus
lists the “official” phrasing for a key word topic and variations on the descriptor // most underused
