# PSYC 217- Final Exam Guide - Comprehensive Notes for the exam ( 59 pages long!)Premium

59 pages187 viewsWinter 2014

Department

PsychologyCourse Code

PSYC 217Professor

Catherine RawnStudy Guide

FinalThis

**preview**shows pages 1-3. to view the full**59 pages of the document.**UBC

PSYC 217

Final EXAM

STUDY GUIDE

###### You're Reading a Preview

Unlock to view full version

Subscribers Only

Only page 1 are available for preview. Some parts have been intentionally blurred.

Subscribers Only

###### You're Reading a Preview

Unlock to view full version

Subscribers Only

Only page 1 are available for preview. Some parts have been intentionally blurred.

Subscribers Only

Probabilistic Reasoning & The Role of Chance

PROBABILISTIC REASONING

- Probabilistic trend: more likely than not but does not hold true in all cases

- Probabilistic prediction = numerical, real prediction (cannot be given with precision)

- Peso-Who “tatistis

• Thee ill also e a peso ho goes against the strongest of trends

• Eg. smoking some will smoke everyday and not get cancer; others do

Psychological Research on Probabilistic Reasoning

- Insufficient Use of Probabilistic Information e do’t ko ho to use it

- Failure to use sample-size information (smaller samples will always generate more extreme values) sampling

error (assuming that the smaller sample will represent a larger population)

- Gale’s Fallacy: tendency for ppl to see links b/w events in the past and future when the two are really

independent we tend to see patterns

CHANCE

- Explaining Chance: Illusory correlation & the illusion of control

• Illusory correlation: he ppl thik the’e seeig co-occurrences feuetl he the’e eall ado

(they see structure where there is none)

• Illusion of control: tendency to believe that personal skills can affect outcomes determined by chance

• Just-world hypothesis: ppl tend to believe that they live in a world in which ppl get what they deserve

- Chance & Psychology

• Coincidence: occurrence of related events due to chance (no other explanation is necessary)

• Oddmatch: two events whose co-occurrence strikes us as odd/strange (can happen if wait long enough)

• Personal coincidences: same birthdays, remembering coincidental events

- Accepting Error to Reduce Error: Clinical vs. Actuarial Prediction:

• Actuarial prediction: predictions based on group trends derived from statistical records (eg. aggregate)

o Predicts the same outcome for all individuals sharing a certain characteristic

o Economics, HR, criminology, medical sciences, etc

o Is superior to clinical prediction in accounting for human behaviour

• Clinical prediction: practitioners claim to be able to go beyond group predictions and to make accurate

predictions of outcomes of particular individuals

o Cog psyc, developmental psyc, etc

In psychological science, the goal is truth:

- Was result due to chance? Or did it really represent truth?

• Analyze data using statistics

• How likely is it that our result is due to random error?

- Aoids elig o hua’s iased thikig

find more resources at oneclass.com

find more resources at oneclass.com

###### You're Reading a Preview

Unlock to view full version

Subscribers Only

Only page 1 are available for preview. Some parts have been intentionally blurred.

Subscribers Only

- Need to think about probability (likelihood of the occurrence of some event/outcome)

POPULATION VS. SAMPLE

- Population level in the Smarties factory, how many of each colour are made?

• Distribution of Smarties colours at the level of the Smarties factory

- Probabilistic trend proportion of each colour of Smarties in the pop of smarties at the factory

• The smaller the sample, the less likely the pop will balance out

- Random sample In any fun sized packet of Smarties, how many of each colour are there?

*As our sample size increase, the pattern in the sample will better represent the truth in the population

Samples are imperfect signals of overall probabilities

- Probabilistic trend:

• Overall, what is the true effect?

o Trying to get true effect eg. proportion of diff colours (correlation, diff b/w two group means)

• Does’t eed to e tue i ee ase fo it to be true generally

• Eg. Regression line (line through scatter plot)

o Dots ost of the tie do’t fall o the staight lie aiatio)

o Does’t ea oeall elatioship is’t tue

- Random sample:

• Take random sample from pop to estimate true effect

• As sample size decreases, estimate is more likely to be wrong

*the smaller the sample the more likely it is to be weird

Why do we need to bother using statistics? Why cant we just look for trends and call it like we see it?

We like to e ight, ut e’e ofte ot

- Focus on (random) rare events

• Fail to use probabilistic info when making judgments

• eg. peso-ho statistis – I ko a peso ho…

• Poailisti teds do’t eed to appl to ee ase to auatel epeset the hole

• Focusing on the one rare event

- See patterns in randomness

• See trends in events that are really due to chance

• Look for patterns in coincidences (eg. Bible code predicting important dates)

• Gale’s falla

o Tendency to see links b/w past & future events at times when past events are completely

independent from future events streak

o Independent events do not predict future

o I’ due fo a i!

find more resources at oneclass.com

find more resources at oneclass.com

###### You're Reading a Preview

Unlock to view full version

Subscribers Only

#### Loved by over 2.2 million students

Over 90% improved by at least one letter grade.