Standard Binary Logistic RegressionLogistic regressionLevel of measurement requirementsDummy-coding in SPSS - 1Dummy-coding in SPSS - 2AssumptionsNumerical problemsSample size requirementsMethods for including variablesComputational methodOverall test of relationshipOverall test of relationship in SPSS outputRelationship of Individual Independent Variables and Dependent VariableInterpreting individual relationships - 1Interpreting individual relationships - 2Strength of logistic regression relationshipEvaluating usefulness for logistic modelsComparing accuracy ratesComputing by chance accuracyOutliersStrategy for OutliersThe Problem in BlackboardSlide 23Marking the Statement about Level of MeasurementThe Statement about OutliersRunning the standard binary logistic regressionSelecting the dependent variableSelecting the independent variablesDeclare the categorical variables - 1Declare the categorical variables - 2Declare the categorical variables - 3Declare the categorical variables - 4Specifying the method for including variablesAdding outliers to the data set - 1Adding outliers to the data set - 2Requesting the outputDetecting the presence of outliers - 1Detecting the presence of outliers - 2Detecting the presence of outliers - 3Detecting the presence of outliers - 4Detecting the presence of outliers - 5Running the model excluding outliers - 1Running the model excluding outliers - 2Running the model excluding outliers - 3Running the model excluding outliers - 4Running the model excluding outliers - 5Running the model excluding outliers - 6Running the model excluding outliers - 7Running the model excluding outliers - 8Running the model excluding outliers - 9Accuracy rate of the baseline model including all casesAccuracy rate of the revised model excluding outliersMarking the statement for excluding outliersThe statement about multicollinearity and other numerical problemsChecking for multicollinearityMarking the statement about multicollinearity and other numerical problemsThe statement about sample sizeThe output for sample sizeMarking the statement for sample sizeThe overall relationship between the dependent and independent variablesThe output for the overall relationshipMarking the statement for overall relationshipThe statement about the relationship between education and computer useOutput for the relationship between education and computer useMarking the statement for the relationship between education and computer useThe statement for the relationship between poor health and computer useOutput for the relationship between poor health and computerMarking the statement for the relationship between poor health and computer useThe statement for the relationship between fair health and computer useOutput for the relationship between fair health and computer useMarking the statement for the relationship between fair health and computer useThe statement for the relationship between good health and computer useOutput for the relationship between good health and computer useMarking the statement for the relationship between good health and computer useThe statement for relationship between socioeconomic index and computer useOutput for the relationship between socioeconomic index and computer useMarking the relationship between socioeconomic index and computer useThe statement for the relationship between sex and computer useOutput for the relationship between sex and computer useMarking the statement for the relationship between sex and computer useStatement about the usefulness of the model based on classification accuracyComputing proportional by-chance accuracy rateOutput for the usefulness of the model based on classification accuracyMarking the statement for usefulness of the modelStandard Binary Logistic Regression: Level of MeasurementStandard Binary Logistic Regression: Exclude OutliersStandard Binary Logistic Regression: Multicollinearity and Sample SizeStandard Binary Logistic Regression: Overall RelationshipStandard Binary Logistic Regression: Individual RelationshipsStandard Binary Logistic Regression: Classification AccuracySlide 1Standard Binary Logistic RegressionSlide 2Logistic regressionLogistic regression is used to analyze relationships between a dichotomous dependent variable and metric or non-metric independent variables. (SPSS now supports Multinomial Logistic Regression that can be used with more than two groups, but our focus now is on binary logistic regression for two groups.)Logistic regression combines the independent variables to estimate the probability that a particular event will occur, i.e. a subject will be a member of one of the groups defined by the dichotomous dependent variable. In SPSS, the model is always constructed to predict the group with higher numeric code. If responses are coded 1 for Yes and 2 for No, SPSS will predict membership in the No category. If responses are coded 1 for No and 2 for Yes, SPSS will predict membership in the Yes category. We will refer to the predicted event for a particular analysis as the modeled event.Predicting the “No” event create some awkward wording in our problems. Our only option for changing this is to recode the variable.If the probability for group membership in the modeled category is above some cut point (the default is 0.50), the subject is predicted to be a member of the modeled group. If the probability is below the cut point, the subject is predicted to be a member of the other group.For any given case, logistic regression computes the probability that a case with a particular set of values for the independent variable is a member of the modeled categorySlide 3Level of measurement requirementsLogistic regression analysis requires that the dependent variable be dichotomous.Logistic regression analysis requires that the independent variables be metric or non-metric. The logistic regression procedure will dummy-code non-metric variables for us. For logistic regression, we will use indicator dummy-coding, rather than deviation dummy-coding since I think it makes more sense to compare the odds for two groups rather than compare the odds for one group to the average odds for all groups.If an independent variable is ordinal, we can either treat it as non-metric and dummy-code it or we can treat it as interval, in which case we will attach the usual caution.Dichotomous independent variables do not have to be dummy-coded, but in our problems we will have SPSS