Stage 2 Determine Positive Deviants Technical Terms Units having similar structural factors that are beyond the level of the individual and are generally common to a group Homologues Contextual factors that are above the level of the individual and the social network includes factors that vary as a function of geographic unit The variable that is the predictor of a change in or effect on the dependent variable Structural variables Independent variable The outcome variable being tested and measured and is dependent on the independent variables Dependent variable Proxy Ground truth data Outliers Data collected by direct observation as opposed to data provided by inference i e proxy In statistics an outlier is an observation that differs significantly from other observations Cut off point The limit used to distinguish positive deviants from non positive devaints Univariate multivariate outlier detection In univariate outlier detection techniques outliers are identified from a single dimension or measure while in multivariate techniques outliers are identified across multiple dimensions This section outlines the different steps of Stage 2 Determine Positive Deviants which starts with performance measurment followed by the homogeneous grouping and positive deviant identification and finally the preliminary validation of the identified positive deviants Section overview An indirect measure of the desired outcome which is strongly correlated to that outcome 53 54

Vorschau DPPD Handbook Seite 28
Hinweis: Dies ist eine maschinenlesbare No-Flash Ansicht.
Klicken Sie hier um zur Online-Version zu gelangen.