Week 9 and 10 Flashcards
What is descriptive statistics?
Obtained through descriptive statistical techniques that reduce data to manageable proportions.
What are some descriptive statistical techniques?
Measures of central tendency (mean, median, mode) Measures of variability (Standard deviations, range) Correlation Techniques (Scatter plots)
What are inferential statistics?
Statistical details that combine math and logic to test hypotheses about a population with sample data.
What are the levels of measurement?
Nominal, ordinal, interval, ratio. (Low to high)
What is nominal measurement?
Variables or events that are classified into categories. Ie. Gender, hair colour, marital status etc. Can be considered dichotomous or categorial.
What is ordinal measurement?
Reveals relative ranking of variables or events. Ie. Ranking, intervals not necessarily equal.
What is interval measurement?
Events or variables that are ranked on a scale with equal intervals between numbers.
What is ratio measurement?
Events or variables are ranked on scales with equal intervals and absolute intervals. Usually only achieved in physical sciences.
What is frequency distribution?
The number of times each event occurs is counted, data organized into groups then frequency is reported.
What is the mode?
The most frequent score.
What is modality?
The number of modes.
What is the median?
The middle score.
What is the mean?
The mathematical average of all the scores, used with interval/ratio level data.
What is a normal curve?
Unimodal symmetrical, mean, median, and mode are all the same.
What are measures of variability?
Statistical procedures that describe the level of dispersion in sample data.
What is range?
The distance between the highest and lowest score.
What is semiquartile range?
The range of the middle 50% of the scores. More stable than the overall range because it is less likely to be changed by a single extreme score.
What is a percentile?
Represents the percentage of scores that a given score exceeds.
What is a standard deviation?
Most frequently used measure of variability and based on the concept of a normal curve.
What is the z-score?
Used to compare measurements in a standard unit. Each score is converted to a Z score, to compare.
What are the 2 purposes of inferential statistics?
To estimate the probability that statistics found in the sample accurately reflect the population parameter and to test hypotheses about a population.
What is a parameter?
A characteristic of a population.
What is a statistic?
A characteristic of a sample.
What do inferential statistics allow researchers to do?
Allows them to make statements about the larger population from studying the sample.
What is the scientific hypothesis?
What the researchers believe the outcome of the study will be. H1.
What is the null hypothesis?
The hypothesis that indicates no statistical difference between groups. NO RELATIONSHIP EXISTS.
What is probability?
The event’s long run frequency in repeated rails under similar conditions.
What is sampling error?
The tendency for statistics to fluctuate from one sample to another.
What is the standard error of mean?
The standard deviation of a theoretical distribution of sample means.
The _______ the standard error the _____ variable the sample means and the ______ accurate those means as estimates of the population value.
Smaller; less; more.
What is a type 1 error?
The researcher’s incorrect decision to reject the null hypothesis. The researcher has found results that are statistically significant, when in fact they are not, and the researcher has accepted the alternate hypothesis.
What is a type 2 error?
(Beta) The results from the sample lead to the failure to reject the null hypothesis, when the null hypothesis is in fact false. The researcher fails to realize there is statistical differences in the data.
What is power?
The conditional prior probability that the researcher will decide correctly to reject the null hypothesis, when it is actually false. When power is increased, type 2 error is decreased and vice versa.
What is the relationship between power and beta?
They are complementary and sum to 1.00.
What type of error is considered more serious and why?
Type 1 errors are more serious because patient care can potentially be effected, due to the researcher declaring differences when their are infact none.
What is the level of significance/Alpha?
the probability of making a type 1 error. Set before the data is collected. Conditional probability because the null hypothesis is assumed to be true.
What is the minimum level of significance accepted in nursing research?
0.5.
What is the p-value?
Probability value. It is the probability of obtaining a test statistic like a mean. Calculated from the sample data and considered the exact level of significance.
If p ____ alpha, the null hypothesis is rejected.
Less than.
What is nonparametric tests of significance?
Not based on population parameters. Assumptions are less restrictive. Usually applied when the variables have been measured on a nominal or ordinal scale.
What are nonparametric statistics?
Best used when the data cannot be assumed to be at the interval level of measurement, or when the sample size is too small.
k
k
What is logistic regression?
Relationships between multiple independent variables and a dependent variable that is binary, ordinal or polynomial are analyzed.
What is the t statistic?
Reflects whether two groups means are different. Variables must be measured using interval or ratio level. Demonstrates that differences exist between groups.
What is the analysis of variance (ANOVA)?
Test similar to t test, used to test whether group means differ. Accounts for the variation between groups within groups. Used with 2 or more groups. Usually performed with F tests, rather than multiple t tests.
What is a multivariate analysis of variance (MANOVA)?
USed to determine differences in group means, but with more than one dependent variable.
What is a post hoc analysis?
Pairs of means in the main effects and interaction effects are compared to determine whether they are statistically different.
What is an analysis of covariance (ANCOVA)?
Used when groups are different at baseline or beginning. Entails measuring differences among group means and helps researchers equate the groups under study on an important variable.
What is the chi-square (x2) statistic?
A nonparametric statistic used to determine whether the frequency in each category is different from what would be expected by chance. If x2 is high enough, null hypothesis would be rejected.
What is correlation?
The degree of association between 2 or more variables.
What is Pearson’s correlation coefficient?
Calculation reflects the degree of relationship between two interval variables. Can range from -1.0 to 1.0.
What is a weak correlation? What is a strong correlation?
0 to 0.2 very weak, 0.2 to 0.4 weak, 0.4-0.6 moderate, 0.6-0.8 strong. 0.8-1.0 very strong.
When is the phi coefficient used?
Can be used to express relationship between nominal variable and interval variable, point-biserial correlation is used.
What is multiple regression?
Happens when researchers are interested in understanding more than just a relationship between variables. The relationship between one dependent variable at the interval level, and several independent variables is measured.
What is the confidence variable?
Range of values, based on a random sample, often described with measures of central tendency and measures of association, provides nurse with measure of precision or uncertainty about sample findings.
What is the odds ratio?
Communicates that one event is likelier to occur than other events.
What is a systematic review?
Process whereby the investigators evaluate all relevant studies, published and unpublished, on the topic.
X increases, Y increases, what is the type of correlation?
Direct or positive, positive interger.
X decreases, Y decreases, what is the type of correlation?
Direct or positive, positive interger.
X increases, Y decreases, what type of correlation?
Indirect or negative, negative interger.
X decreases, Y increases, what type of correlation?
Indirect or negative, negative intereger.
What is multiple regression?
One dependant variable, 2 or more independent variables. Trying to predict 1DV.
what is linear regression?
Using one IV to predict 1 DV.
What does the R^2 value tell you?
Proportion of variability in the DV explained by the IVs.
