PSYB04H3 Lecture Notes - Lecture 12: Publication Bias, Scientific Literature, Meta-Analysis
PSYB04
LEC 12
Chapter 14: Replicability, Generalization, and the Real World
To Be Important, a Study Must Be Replicable
● Researches always consider whether they will get the same result if they do the study again → if it
is replicable
● Can the result be replicated
● Gives a study credibility; crucial part of scientific process
→ Replication Studies
● Even when original study has statistically significant result, researchers still need to conduct the
study again to find out if the result is replicable
● Direct replication
○ Researchers repeat an original study as closely as they can
○ See whether the original effect shows up in newly collected data
○ Direct replication can never replicate the first study in every detail
● Conceptual Replication
○ Study the same research question but use different procedures
○ Variables in the study are the same but the procedures for operationalizing the variables
are different
● Replication-Plus-Extension
○ Researchers replicate their original study but add variables to test additional questions
○ Add another situational variable
● Replications by Independent Researchers
○ Failure to replicate a finding in a completely diff lab raises the possibility that the original
effect can be obtained only in very specialized conditions
→ Replication, Importance, and the Weight of the Evidence
● A single study is important only if it can be replicated
→ Meta-Analysis: What Does the Literature Say?
●
Scientific literature →
consists of a series of related studies, conducted by various researchers,
that have tested similar variables
○ Several studies of a particular topic]
●
Meta-analysis →
mathematically averaging the results of all the studies that have tested the same
variables
○ See what conclusion that whole body of evidence supports
○ Researchers collect all possible examples of a particular kind of study
○ Average all the effect sizes to find an overall effect size
○ Publication bias in psychology → significant relationships are more likely to be published
than null effects
find more resources at oneclass.com
find more resources at oneclass.com