Research Design Decision Tree

Mind Map by , created over 6 years ago

Mind Map on Research Design Decision Tree, created by josman9 on 07/09/2013.

Tags No tags specified
Created by josman9 over 6 years ago
Attachment - Psychology - Flash Cards
Megan Price
Cell Transport
Elena Cade
The Periodic Table
Romeo & Juliet Quotes
Lucy Hodgson
Core 1.8 Metals
T Andrews
Elements, Compounds and Mixtures
Unit 1 Sociology: Family Types
Biology AQA 3.1.5 The Biological basis of Heart Disease
A Level: English language and literature technique = Dramatic terms
Jessica 'JessieB
Conceptos Generales De Robótica
fede ramos
Research Design Decision Tree
1 How Many Variables?
1.1.1 Scale of Measurement? Nominal Info about Distribution? Central Tendency: tables for modal value Distribution: tables for frequency of modal value or class Frequencies: tables for relative and absolute Ordinal Info about Dispersion? Central Tendency: tables for median Dispersion: need the inter-quartile deviation Frequencies: tables for relative and absolute Interval Info about Distribution? Symmetry: calculate skewness Dispersion: Central Tendency: Skewed: compute the mean and median Symmetric: compute the mean Normality: Normality: Kolmogorov-Smirnov one-sample test, Lilliefors extension of Kolmogorov-Smirnov test, Chi-square goodness-of-fit test, the Jarque-Bera test, D'Agnostino-Pearson K-squared test, Shapiro-Wilk test. Skewness & kurtosis: D'Agnostino-Pearson K-squared Jarque-Bera Frequencies:tables for relative and absolute. Consider requesting n-tiles Peakedness: compute the kurtosis of a variable To test departures from normality: for N greater than 1000, refer the critical ratio of the kurtosis measure to a table of the unit normal curve; for N between 200 and 1000, refer the kurtosis measure to a table for testing kurtosis; for N less than 200, use Geary's criterion.
1.2.1 Scale of Measurement? 1 Interval, 1 Nominal Is interval variable dependent? YES: measure of strength or test of significance? Test of significance: assuming homoscedasticity across levels of ind. variable, perform an analysis of variance and F-test for significance With no homoscedasticity across levels of ind. variable, use ANOVA. For hypothesis testing use the Welch statistic, the Brown-Forsythe statistic, or the t-test Measure of strength: Use the ANOVA, and Omega Squared Intraclass Correlation Coefficient Kelley's Epsilon Squared NO: ANOVA to perform an analysis of variance 1 Nominal, 1 Ordinal compute the Friedman test and probability of chance occurrence. Use Freeman's coefficient of differentiation, theta 1 Interval, 1 Ordinal If ordinal is based on an underlying normally distributed interval variable, use Jaspen's Coefficient of Multiserial Correlation Both Nominal Both variables 2-point scale? YES: What will be measured? Symmetry: Use McNemar's test of symmetry; it is equivalent to Cochran's Q Covariation: use Yule's Q Phi NO: At least one is not a 2-point scale and one is considered an independent variable Statistic based on number of cases in each category use Goodman and Kruskal's tau b Statistic based on number of cases in modal categories calculate the asymmetric lambdas A and B Both Ordinal Distinction between dependent & independent variables? YES: use Somer's d for 2 ordinal variables NO: What do you want to measure? Agreement: no applicable statistic, but data may be transformed to ranks and r or Krippendorff's r used Covariance: depending on if the ranks are treated as interval scales, use Kendall's tau-a, tau-b, tau-c Goodman and Kruskal's gamma, Kim's d, or Spearman's rho (rs) Both Interval Distinction between dependent & independent variables? YES: looking for linear relationship? YES: use the F-test, also computed by Regression NO: Curvilinear relationships- use the F-test, computed by Regression, equal to t-squared, for each coefficient NO: looking for equal means on both variables? YES: calculate the t-test for paired observations NO: treat the relationship as linear What do want to measure? Agreement: penalty without same distribution? YES: Robinson's A or the intraclass correlation coefficient. The test is the F-test. NO: Use Krippendorff's coefficient of agreement Covariance: Use Pearson Product-Moment r (correlation coefficient), Biserial R, or Tetrachoric r depending on how many of the variables are dichotomous
1.3 More than 2 Variables
1.3.1 [Didn't Learn This]

Media attachments