Textbook Notes (280,000)
CA (160,000)
UW (6,000)
PSYCH (1,000)

PSYCH291 Chapter Notes -Meta-Analysis, Falsifiability

Course Code
John Holmes

of 1
Misunderstanding of breakt/ headlines: implication tht all problems in sci solved when sgl, crucial expmt
completely decides issue, or tht theoretical advance is result of sgl
critical insight tht overturns all previous knowledge
The Connectivity Principle
Note: not arguing breakt/ instances dN occur
i.e. Albert Einstein’s “Theory of Relativity”
Connectivity principle = new theory in sci must make contact w/ previously established empirical facts
For new theory to be considered an advance, must not only explain new facts but acc’ for old ones
A Consumers Rule: Beware of Violations of Connectivity
Breakt/ model (aka “Einstein Syndrome”) implies new discoveries violate principle of connectivity
- benefits pseudosci & bogus theories, which can claim “newness”
- make theory: 1) unfalsifiable; and 2) dismiss previous data as irrelevant (TF) no connectivity
The Great-LeapModel vs the Gradual-Synthesis Model
Many areas in sci advanced not by sgl, sudden breakt/s but by series of fits & starts
More often multiple expmt’ns & ruling outs + criticisms before any advancement
- “Great-Leap”s like Einstein happen, but rare
Converging Evidence: Progress Despite Flaws
Principle of converging evidence = aka converging operations; eval’n tool; useful for lay consumer of sci
info; tells us to examine pattern of flaws running t/ research literature
b/c nature of pattern can support/undermine conclusions
When evidence from wide range of expmts, each flawed w/ tm’s of diff s&w’s, pts in ~ direction, then
evidence has converged
Research highly convergent when series of expmts consistently supports given theory while collectively
eliminating most important competing explanations
Converging Evidence in Psychology
Conclusions often based on principle of converging evidence
Counterweight to warnings against over-interpreting tentative knowledge
Scientific Consensus
Rule: when eval’g empirical evidence in psych, think in terms of scientific consensus rather than
breakthroughin terms of gradual synthesis rather than great leap
Methods and the Convergence Principle
Relative balance among methodologies used desirable b/c various classes of research tm’s have diff s&w’s
The Progression to more Powerful Methods
Case studies corr’l studies expmts w/ manipulated variables
A Counsel Against Despair
Use of meta-analysis