Study Guide for Test 2

Dr. Gass

HCOM 308

Summer 2009

Study Guide for= Test 2

Chapter 8 <= /o:p>

*perameter (whole population), p.144=

**sample (subset of a population), p. 145

*data (records, observations, or measurements), p. 146

***descriptive statistics, (a.k.a. summary statistics), p. 146

single numerical indicators that describe or summarize data sets

***measures of central tendency (averages), pp. 147-157

****mean or arithmetic average, pp. 147= -148

&= nbsp; ***most sensitive, precise measure of central tendency

&= nbsp; ***affected by extreme scores, e.g., outliers

&= nbsp; ***may skew data

&= nbsp; ***most useful for statistical analysis

&= nbsp; ***median, pp. 148-149

&= nbsp; ***physical middle point of a distribution, middle value in a set of scores<= /span>

&= nbsp; ***as many scores above as below the median

&= nbsp; ***often reported along with the mean

&= nbsp; ***less subject to skewing by extreme scores

&= nbsp; **mode pp. 149-150

&= nbsp; *used infrequently in social science research

&= nbsp; *bimodal and multimodal distributions

&= nbsp; *unaffected by extreme scores, e.g., outliers

***measures of variability (dispersion), pp. 157-165

&= nbsp; **range, p. 158

&= nbsp; difference between highest and lowest scores

&= nbsp; **variance (just the concept, not the formula), pp. 159-160

&= nbsp;

&= nbsp; ****standard deviation (just the concept, not the formula), pp. 160-162

&= nbsp; *average amount of deviation in a distribution

&= nbsp; *most sensitive, precise measure of variability

***displaying descriptive statistics in SPSS, pp. 151-154

Chapter 13<= /o:p>

**pretests and posttests, p. 259

**antecedent variables, p. 259

**intervening variables, p. 259

***random assignment, pp. 260-261, p. 268

****equal likelihood principle

***random assignment matters more than random sampling

*confederates, p. 260, p. 263

**ways of manipulating independent variables

&= nbsp; intact or self-selected groups

&= nbsp; stimulus conditions, p. 263

&= nbsp; stimulus videos, p. 263

&= nbsp; stimulus materials, p. 263

&= nbsp; hypothetical scenarios, role playing, p. 263

&= nbsp; problem solving, decision making

&= nbsp; manipulations must be substantial enough to produce a change, p. 265

****factorial designs, pp. 264-265

&= nbsp; ****number of variables versus number of conditions

&= nbsp; ****2 X 2 design, 2 X 3 X 2 design, etc., p. 264

&= nbsp; **sample size

rule of thumb: 30= -50 subjects per condition

minimum cell size 5-= 10 subjects, p. 265

&= nbsp; **as the number of conditions increase, demand for subjects increases exponentia= lly

**measurement of the dependent variable

&= nbsp; **scales and diagnostic instruments, p. 265

&= nbsp; **behavioral measures, p. 265

&= nbsp; **physiological measures, p. 265

***common experimental designs

&= nbsp; ***Campbell and Stanley’s 3 types of experiments

&= nbsp; *pre-experiments, pp. 273-274

&= nbsp; &nbs= p; one-shot case study

&= nbsp; &nbs= p; one group, pretest-posttest design

&= nbsp; &nbs= p; static group comparison &n= bsp;

**quasi-experiments, pp. 274-276

&= nbsp; &nbs= p; two group, pretest-posttest design

&= nbsp; &nbs= p; time-series design

true experiments<= o:p>

&= nbsp; &nbs= p; ***pr= etest, posttest, control group design, p. 276

&= nbsp; &nbs= p; ***po= sttest only, control group design, p. 277

&= nbsp; &nbs= p; ***So= lomon four-group design, p. 278

Chapter 14<= /o:p>

****sample versus population

***to be generalizable (p. 283) a sample should be:

&= nbsp; ***random

**representative

&= nbsp; **sufficient in size

**sampling frame: comprised of all the available participants in the population

***equal likelihood principle, p. 284

****probability sampling, randomly selecting participants from the population, p. 284<= /o:p>

**central limits theorem, p. 284

**simple random sample, p. 285

**stratified sample, p. 286<= /span>

**cluster sample, p. 287

**systematic sample, p. 287<= /span>

*periodicity, hidden order to

****nonprobability sampling , = based on convenience, availability, p. 288

&= nbsp; ****convenience sample—most common type in social science research, p. 289=

&= nbsp; **volunteer sample, pp. 289-290

&= nbsp; ***purposive sample, p. 290

&= nbsp; **quota sample, p. 290-291

&= nbsp; **network sample, p. 291

replication, pp. 293-294

standard<= /b> normal distribution

     =            **know the underlying assumptions of inferential statistics

     =            **conti= nuously distributed variable

     =            **proba= bility theory

     =            ****are= as under the normal curve

        &= nbsp;       ****know the proportion of cases that are +/- 1 SD, +/- 2 SD, +/- 3 SD from the mean=

        &= nbsp;       ****know where 68%, 95%, and 99% of cases would be

        &= nbsp;       **** significance level, p < .05, p < .01,

difference tests, Ch. 16, 17, 18

****know the symbols for each test we covered in class

****know when and under what circumstan= ces each test should be used

****chi-square test, Ch. 16<= /span>

***cross-tabs or contingency tables

****independent samples t-test, Ch. 17<= o:p>

***one-way ANOVA, Ch. 18

***two-way ANOVA

***factorial analysis of variance (2 X = 2, 3 X 3, etc.)

****know the difference between the num= ber of variables and the number of conditions

****main effects and interaction effect= s

        &= nbsp;       **statistical artifact (chance result)

correlation tests, Ch. 19

****what correlation tests prove

&= nbsp;           &nbs= p; ***direction of correlation

&= nbsp;           &nbs= p; ***degree or strength of association

&= nbsp;           &nbs= p; ***statistical significance

***know the r values for a weak, modera= te, or strong correlation

****bivariate correlations, e.g., two variables

****scatterplot or scattergram

****positive versus negative correlatio= ns

***linear, nonlinear, curvilinear relationships

        &= nbsp;       ****correlation coefficient (know what an “r” value represents)

        &= nbsp;       ****Pearson r

        &= nbsp;       **Spearman rho and Kendall’s tau-b

        &= nbsp;       *Cramer’s V

**correlation doesn’t prove causa= tion

**coefficient of determination (just kn= ow the basic concept)

**spurious relationship

SPSS<= /span>

        &= nbsp;       ***creating a variable

        &= nbsp;       ***values and value labels

        &= nbsp;       ***columns versus rows

        &= nbsp;       ***levels of measurements in SPSS (nominal, ordinal, scale)

        &= nbsp;       entering and saving data

        &= nbsp;       ***select cases; “select if” condition

        &= nbsp;       ***creating a scattergram

        &= nbsp;       ***creating a histogram

        &= nbsp;       ***interpreting various SPSS results, e.g. p < .05, p < .01, p < .001

        &= nbsp;       **performing a chi-square

        &= nbsp;       **performing a t-test

        &= nbsp;       **performing a one-way ANOVA

        &= nbsp;       **performing a Pearson r

        &= nbsp;       **if you see columns of data, you should be able to tell which significance test applies

**if you see a results table, you shoul= d be able to interpret it

            =