Research Methods Exam 3 Review

Term

1 / 90

population

Click the card to flip 👆

Terms in this set (90)

population

the whole universe of people the researcher wishes to study.

sample

a small portion of the population and should represent the population

hypothesis

are assumptions or educated guesses. we test hypotheses using inferential statistics

null hypothesis symbol

Alternative Hypothesis symbol

null hypothesis

its the hypothesis of no difference. It states the independent variable (smart pill is the I.V.) will NOT work...or there will be NO relationship

Alternative Hypothesis

it si the hypothesis of difference, it states the I.V. WILL work ...or there WILL be a relationship.

Directional Hypothesis

we predict the direction scores will change.
(Think of words like: positive, negative, increase, decrease, etc.)

Non-Directional Hypothesis

we do not predict the direction scores will change.

Statistical Significance

it determines the extent to which the differences relationships found may have occurred due to chance. It helps us determine if a Null Hypothesis can be rejected and if an Alternative Hypothesis can be accepted.

the lower the p-value the ______ the significancehigherp= .01 significant or not significant?significantp=.50 significant or not significant?not significantp=.001 significant or not significant?very high significancep=.04 significant or not significant?significantAlphathe level of statistical significance USED to decide whether to rejector not reject the nullp-valuethe level of statistical significance OBTAINED is called the probability value (aka the p-value)the most commonly used Alpha in educational and psychological research is____________..05 A p-value of less than .05 indicates statistically significant results. A p-value of less than .05 means there is less than a 5% chance of making an error.type I errorwe reject the null when the null is true. we conclude that the I.V. works when it really doesn't.type II errorwe retain the null when the null is false. we fail to identify that the I.V. really does work.the type of statistical analysis used depends on:1. the hypothesis or research question 2. the research design 3. the type of sample (within design/between design) 4. the type of data (NOIR- nominal, ordinal, ration data)2 types of samplesdependent samples- within sample and independent samples- between samplenon-parametric tests are used if...**1. the data are categorical which is nominal/ordinal** 2. the data are not normally distributed 3. there is not homogeneity of varianceparametric tests are used if...**1. the data are continuous, so interval data or ratio** 2. the data are normally distributed 3. there is homogeneity of varianceratio dataThe data is ordered with intervals between data values that are equal and contains a true zero point. Think: ZERO MEANS ZERO! Example: having zero dollarsnominal datanaming categories, order does NOT matter.ordinal datathe items ARE ranked in order, the spaces between rankings are not equal.interval datathe data is ordered and the intervals between the data are equal. Example: temperatureAPA stands for...American Psychological AssociationAbstract-summary of article -150 words or lessIntroduction/Lit Review-problem discussed -historical background -trends/gaps -concludes with hypothesis or research questionMethod-participants -instruments: survey, IQ test, etc. -procedures: experiments vs control, etc.results-stats -p-value, etc.discussion-summary -related it to the literature -implications -limitations -ideas for future researchorder of APA format paper-title -abstract -introduction/lit review -method -results -discussion -reference pagestandard deviationit indicates the average deviation from the mean, and how far the scores are spread out around the mean.Nrepresents the total number of scores in a data set.Frequency (f)the number of times a score occurs. The sum of all individual frequencies in a sample equals N.cumulative frequency (cf)the frequency of all scores AT or BELOW a particular score.percentilethe percent of all scores in the data that are at or below the score.the normal distributionthe normal curve is symmetrical. The left half is the same as the right half. 50% of the scores are on the right, the other 50% is on the left. Extreme scores have a low frequency. The farther a score is from the central score the less frequently it occurs.Extreme scores have a _____ frequency.lowThe farther a score is from the central score the ____ frequently it occurs.lessX represents..raw scoresE representsto add up, the sum ofEX representsadd up the raw scorescentral tendencya score that summarizes the location of a distribution on a variable. It indicates where the center of the distribution tends to be located.modethe most frequently occuring score in the data set. Think: Mode/Mostmediananother name for the score at the 50th percentile, it separates the lower 50% from the upper 50%. It is the middle score in a set of score.meanthe score located at the mathematical center of the distribution. It is the average of all the scores.symmetrical distributionthe mean, median, and mode are the same score.measures of variabilityindicate how far the score are spread out. large differences below the scores indicate a large amount of variability. small differences below the score indicate a small amount of variability. The more the scores vary, the harder it is to predict scores.rangethe distance between the two most extreme scores in the distribution.variancewhen calculating the variable and standard deviation the researcher is measuring how much the score are spread around the mean. The variance and standard deviation account for every score in the distribution.when testing the statistical significance of the difference between 2 or more proportions, we must consider:1. our sample -dependent (within) -independent (between) 2. our data -categorical (nominal/ordinal)~ non-parametric -continuous (interval/ratio)~ parametricthe most common non-parametric test used to compare 2 or more proportions are the _________________.chi-square testssymbol for chi-squarewhat words should you look for in the hypothesis to see if they are using chi-square?rate or proportionchi-square tests are used to determine if the difference between 2 proportions (or more) is ___________ different.statistically significantly differentwhen comparing 2 or more proportions with dependent samples (within/paired) the _________ chi-square test is commonly used.McNemar chi-square testwhen comparing 2 proportions with independent samples (between/unpaired) the _____________ chi-square test is commonly used.Pearson chi-square testchi-squares only use ________ data.nominalcorrelation research designis used to describe the relationship between 2 or more variables. You get a: -correlation coefficient -p-valuecorrelation coefficientit ranges from -1.00 to +1.00. The sign indicates the relationship.what is the Pearson Correlation symbolrwhat is the Spearman Correlation symbolrsa positive correlation meansas X increases, Y increasesa negative correlation meansas X increases, Y decreaseswhat does -1.00 indicate?a perfect relationshipwhat does .00 indicate?zero means no relationshipwhat does +1.00 indicate?a perfect relationship.00 to .20 is considered what?very weak.20 to .40 is considered what?moderate.40 to .60 is considered what?medium.60 to .80 is considered what?strong.80 and above is considered what?very strongthe closer to __________ the stronger the correlation1.00when describing the relationship between 2 variables in a correlation we need what ?1. our sample -dependent (paired/within) *remember a correlation is NEVER independent because it is paired* 2. our data -categorical (ordinal and non-parametric) -continuous (interval/ratio- parametric)the Spearman rank order correlation is used when ________________.when both variables are non-parametric (ordinal)What is the most commonly used non-parametric technique for examining relationships between 2 variables. It yields a rs (correlation coefficient) from -1.00 to +1.00.The Spearman Rank Order CorrelationThe Pearson Product Moment Correlation is used:when both variables are parametric (which means interval/ratio)The Pearson Correlation is the most commonly used __________ technique for examining relationships between 2 variables. It ranges from -1.00 to +1.00.parametricthe key words to look for that indicate the Spearman Rank Order Correlation was used are:relationship & rankedis rank nominal or ordinal?ordinalwhen dealing with correlation studies and the key word "relationship" is used, is the sample going to be independent or dependent?dependent, every time. When dealing with relationships- the sample is NEVER independent.the key word "rate" is used for what method?chi-squaredependent samples use which chi-square?McNemar Chi-SquareA research design that consists of a pre-test and post-test group only design is independent or dependent?dependentMcNemar Chi-Square test is parametric or non-parametric?non-parametricwhat key word indicates the use of chi-square?"rate"