## PreparED Study Materials

# Week 5 - PSY 418: Exam 1

Author: **Student **
Professor: **Andreana P Haley**
Term:

Notecards for test 1 - all conceptual

(Be the first to review)

**Front**

**Back**

What makes psychology a science?

We can study solvable problems through two methods: common sense and the scientific method

Systematic Observation and Public verification

Systematic Observation: Information retrieved from senses (i.e. watching a reaction between two people)Public Verification: The replication of results peer reviews, and the spread of information

Scientific Method versus Common Sense

Methods we use to solve scientific psychology questions.Scientific Method: Explains occurrence, has boundary conditions and refutable (can be tested)Common Sense: Irrefutable because no facts or process, pragmatic (what works rather than why), unfounded claim

Operational Definition

Defining things by how they're measured-Communication/classification-Tests hypothesis

Another term for the y-axis

Ordinate

Another term for the x-axis

abcissa

The Law of Parsimony

The idea that the simplest conclusion/explanation of an event is the correct explanation

Epistemology

Study of knowledge

Ways to obtain knowledge

Through authority, faith, mysticism, idealism, rationalism, emperialism

What are the 3 goals of science?

Describe - what's happening?Predict - why might this be happening?Explain - underlying WHY it's happening

Descriptive methods and predictive methods

Descriptive: (watching in order to describe) Naturalistic Observation, laboratoryPredictive: (manipulate in order to see how something reacts) Correlational or quasi-experiment

Structure of a research paper and its content

Intro - -Introduce topic and terms -Summarizes previous literatureMethods- -Materials and procedures and describe study participantsResults --Contains all results of the studyDiscussion--Brief summary of results and explanation-limitations-how results fit with previous studies-conclusion & future directions

Measures of cognitive function

IQEducationAge

Descriptive statistics

The purpose is to DESCRIBE a data set

Population vs. sampleParameter vs. statistic

Population & parameter go together: population is defined as all the individuals of interest and a parameter is a characteristic of the populationSample and statistics: A subset of the population of interest and a statistic is a characteristic of the sample (CAN OFTEN BE A PREDICTOR OF THE POPULATION)

Central tendencies and measures

Data sets centered around certain point with a certain degree of spread.Mean- usually best, average of all data, can be moved due to extremesmedian- one number the middle of data, only based on ONE number so may not be best measuremode- most frequent number recorded, can make tails and abnormalities more often but not always best measure of center

What does the standard deviation do to a distribution

smaller deviation- narrows the graphed distributionlarger - widens the graphed distribution* s=1 is STANDARD NORMAL DISTRIBIUTION

Qualities of the standard normal distribution and SNC

-Total area under SNC = 1-Extends indefinitely but never touches x-symmetric about 0-Almost all data lies under +/- 3 SD-mean is 0 and SD is 1 and said to have the "standard normal distribution" the correlated curve is the standard normal curve

Reasons for statistical analysis

Because previous answers were varied

Purpose of statistics

It simplifies and condenses a complex model

What do histograms tell about a data set?

They tell the distribution of a data sample. Good for discrete variables because measured in whole units

Frequency Polygons

Best for continuous measure of data

Bar graphs

Best for categorical distribution because it separates them and measures them independently

Advantages of all measures of CT

Mean - best measure of CT because includes ALL numbersmedian- true midpoint and stability because extremes won't affect too muchmode- Indicates shape and abnormalities of a distribution

Disadvantages of measures of CT

mean - can be influenced by extremesmedian - only based off very few scores, not entire distributionmode - may not be representative of the center at all (i.e. skews)

When to use MCTs

If variance is small, MCTs are goodIf variance is large, Measures of Variance would most likely be more representative of distribution

Measures of Variance

Range - based on two numbers. Difference between largest and smallest numberStandard deviation - average deviation from the mean, includes all scores. Simplest measure of parameter

Normally Distributed Variable

DISRIBUTION HAS THE SHAPE OF A NORMAL CURVEPercentage of all possible observations in a specific range also the corresponding area under the curve expressed as a percent

Difference between variance and standard deviation

Variance = s^2 is the sum of the squared errors (x-xmean) divided by n-1Standard deviation = s ^^ also square rooted in addition

What is a normal curve?

A bell-shaped curve with the mean at the centerextends infinitelyarea under curve = 1>99% of area falls under +/- 3 SDs

What is the purpose of a z score and a z test?

Z-score: Compares scores to each other and to the mean of the distribution (puts them all on the same scale): "How far above or below mean you are in relation of the sample to the population" Also allows us to compare data from one distribution to another because of the equal scale it creates.Z-test: ESTIMATES RELIABILITY OF JUDGMENTS MADE, Compares sample mean to the population: "does this sample belong to this population or a different one?"

What is the standardized version of x?

It's the z score of x.

How to determine % or P(x)

1. Sketch the normal curve2. Shade region of interest/ x-values3. Compute z-scores for x-values4. Use table to compute are under SNC

Qualities of a normally distributed variable

1. > 68.26% of all possible observations lie within one standard deviation on either side of the mean (between µ - s and µ + s). 2. > 95.44% of all possible observations lie within two standard deviations on either side of the mean (between µ - 2s and µ + 2s). 3. > 99.74% of all possible observations lie within three standard deviations on either side of the mean (between µ - 3s and µ + 3s).

Reason for and logic of inferential statistics

To interpret data and predict reliability. Makes inference beyond data. Provides estimate of reliability

Probabilities

Inferences: can only estimate the likelihood of an occurrence

Hypothesis testing - directionality and errors

Alternate hypothesis: what we WANT to prove Null hypothesis: opposite of what we want to proveDirectional: we predict what will happen Nondirectional: we predict a change but don't know which way it will sway.Type 1 error- If we say there is an effect when there isn'tType 2 error- If we say there is no effect when there is

Statistical Power

Ability to detect effect where one exists purpose of n. The higher n is, the lower the SD of sampling distribution is, so the higher the z score

Variable Distribution vs. sampling distribution

Variable- mean of scores of entire populationsampling- distribution of means of sample of population

Critical regions & tails

Regions defined by alpha. Number of tails dictates:-size of critical value-directionality tells you if you use 1 or 2 regions and where to place it

T-test

used when there is no SD and z test is impossible -can be small samples

Single T-test assumptions

Comparing mean of sample to populationAssumptions--Random selection of observations: equal opportunity to be selected-Independent observations: only attributed once-Normally distributed-Interval/ratio data

Independent t-test & assumptions

Comparing means of two groups to see if they're part of the same population. Difference between 2 group means relative to standard errorAssumptions:-random sampling -independent observations-normal distribution-homogeneity of variance -interval/ratio data