Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Thinking Like a Psychologist Part III Evaluating the Research • Each research study makes its own contribution • More importantly, how does each study relate to the others and to the theory being tested? • Connectivity = old + new – A good adaptation to a theory not only explains the new findings but accounts for previous findings Evaluating the Research • Don’t jump to conclusions!!! – Media is notorious for doing this • Previous findings are not to be discarded because of one new finding. – Replication, replication, replication – Evaluate the methods used • Is someone trying to say correlation = causation? Evaluating the Research • Converging Evidence – Multiple research studies reveal similar findings. • Convergent Validity – Measures that are predicted to be related, are. • Diverging Evidence – Research that is contrary to the typical findings. • Divergent Validity – Measures that are predicted to be unrelated, aren’t. Evaluating the Research • When designing research: – Read the available literature – Examine the shortcomings (addressed and unaddressed) – Review researcher recommendations – Note the variables examined • Is there another angle to examine? – Note the predominate findings and explanations Evaluating the Research • Meta-analysis – Compilation of multiple research findings from multiple studies examining the same research question • • • • Find all relevant articles Evaluate the articles for inclusion Code the data to be included Examine the outcomes Evaluating the Research • Events and phenomena typically do not have one cause. • Although a research study may only examine one or two variables, that does not mean those the only ones involved. – Multiplicity of Causation – Interaction Effects Evaluating the Research • Sample size is important in evaluating the results – n = 25 vs. n = 300 – A smaller sample size is not as representative of the population – Increasing sample size and number of samples, decreases error. Samples and Populations • Central Limit Theorem – As sample size increases, the distribution approaches normal – Mean of the sample means is equivalent to the population mean – Standard deviation of the sample means is the standard error The math: 9 + 5 + 12 + 22 + 19 + 16 + 13 + 10 Sample1 M=5 Sample 3 M = 12 Sample 4 M=9 Sample 8 M = 19 =106 Population X µ = 15 Sample 7 M = 16 Sample 6 M = 22 Sample 5 M = 10 Sample 2 M = 13 106/8 = 13.25 Central Tendency • Mean, Median, Mode – Mean • All of the scores summed and divided by the number of scores (M = Σx/n); a.k.a. the average – Median • The values separating the upper half of the scores from the lower half. – Mode • The score that occurs most frequently