PSC 202 1st Edition Lecture 11 Outline of Last Lecture I. Case StudiesII. Structured, focused comparisonIII. Process tracing Outline of Current Lecture I. Review Current LectureI. Measurement- measurement error 1. systematic error: the measurement tool consistently mismeasures what we want it to measure; tool is measuring something else2. random error: the measurement tool produces errors that are unpredictable and erratic- assessing measurement quality 1. reliability: extent to which a measurement tool measures the intended concept consistently2. validity: extent to which a measurement tool measures the intended concept with accuracyII. Variables- a measure of a concept- variables have units and can take on one of at least 2 values- must be able to vary- measurement levels: nominal (categorical), ordinal (ordered categories), interval (exact differences)- index variable: a variable that is constructed by adding together a set of variables (average)- data collection: when recording data, we commonly use numeric codes to represent the values of a variable, and most data sets come with a “codebook” that lists the values that go with each codeThese notes represent a detailed interpretation of the professor’s lecture. GradeBuddy is best used as a supplement to your own notes, not as a substitute.- distribution of a variable: the relative frequencies of the different values of a variablein some sample of cases. Often depicted in a histogram or density plot III. Central tendency, distribution &dispersion- central tendency: what is the “center”- dispersion: degree of variation across possible values- central tendency measures1. mode: most common value2. median: middle point3. mean: arithmetic average- nominal  modeordinal mode, medianinterval mode, median, mean- dispersion: how evenly are the cases spread across all possible values of the variable?1. high dispersion: cases clearly spread evenly across values, no clear mode OR multimodal (mode and median split apart, Mean is not typical due to skewness or multi-modality.2. Low dispersion: cases clustered around the mean or median value, AND single peaked (unimodal), mean is typicalSkew: asymmetrical distribution of cases. Mean>median mean<medianIV. Hypotheses - Pollock’s formula: In a comparison of [units of analysis], those having [one value of the independent variable] will be more likely to have [one value of the dependent variable] than will those having [a different value on the independent variable].- elements of a good hypothesis1. a comparison between two variables2. the relationship between the variables is clearly specified and measurable3. unit of analysis is clear4. it is testable and falsifiable V. Case study vs. Large N sample- case study is more specific, detailed and accurate, but not generalizable- large N more generalizable but lack specific details; measures average