630 chp 8-14.docx - Ch 8 Visual Analysis of Graphic Data...

• 12

This preview shows page 1 - 3 out of 12 pages.

Ch. 8 Visual Analysis of Graphic DataVisual Analysis of Graphic DataVisual analysis involves systematic procedures used to evaluate specific characteristics ofdata patterns and evaluate the presence of a functional relation. Visual analysis can be used to evaluate data of individuals or small groups depending on the unit of analysis specified in the research questionSecond, it is a dynamic process in that data are collected repeatedly, graphed as they arecollected and analyzed frequentlyThird, visual analysis focuses on analysis of individual data patterns, thereby facilitating individualization, rather than group- based generalizationsVisual analysis of graphic data permits discovery of potentially interesting findings that may not be directly related to the original research question or program objectiveUsing Visual Analysis to Identify Behavior Change and Functional RelationsFormative visual analysis is conducted within and across conditions to identify behavior change during the course of the studyBehavior change occurs when data patterns in one condition are different from data patterns in the subsequent, adjacent condition for the same variable(s)Summative visual analysis is conducted following study completion, across multiple opportunities to demonstrate behavior change to determine whether a functional relation exists between the independent and dependent variablesAdjacent conditions are when data in one condition differ from what is predicted basedon the preceding condition, behavior change is demonstratedFormative Visual Analysis: Within Condition AnalysesWithin condition visual analyses are conducted to discern patterns within a single condition during a study Beginning with the initial condition – typically baseline – you should look for stability of data across a minimum of at least three/3 to five/ 5 sessions prior to changing conditions The term level refers to the amount of behavior that occurs, as indicated by the ordinatescale value Level is often the characteristic of the highest interest for behavior change, and is generally described as low, moderate, or high Trend is the slope and direction of a data series or the direction data are moving over time (increasing, decreasing, or remaining)Trend direction is referred to as accelerating (increasing in ordinate value over time, decelerating (decreasing in ordinate value over time), or zero celerating (data series is parallel to the abscissaTrend can further be characterized by magnitude and is often described as steepor gradualand paired with paired with direction (e.g., steep accelerating trend or gradual decelerating trend).To increase confidence in functional relations, trend and direction and stability should align with hypothesized data patterns
Variability is fluctuation from one data point to the next and is the opposite of stability; in data with no trend, variability can be summarized as the range of data values within a