Observers may also use rating scales to rate phenomena along a dimension of

Observers may also use rating scales to rate

This preview shows page 24 - 25 out of 55 pages.

frequency, or duration of behaviours or events. Observers may also use rating scales to rate phenomena along a dimension of interest (e.g., lethargic/energetic). Structured observations often involve a sampling plan (such as time sampling or event sampling) for selecting the behaviours, events, and conditions to be observed. Observational techniques are often essential, but observational biases can reduce data quality. Data may also be derived from biophysiological measures, which include in vivo measurements (those performed within or on living organisms) and in vitro measurements (those performed outside the organisms body, such as blood tests). These measurements have the advantage of being objective, accurate and precise. In developing a data collection plan, the researcher must decide who will collect the data, how the data collectors will be trained, and what circumstances for data collection will be In quantitative studies, variables are measured. Measurement involves assigning numbers to represent the amount of an attribute present in a person, using a set of rules; researchers strive to use measures that have good rules that minimize measurement errors. Measures (and the quality of the data that the measures yield) can be evaluated in a psychometric assessment in terms of several measurement properties, most often reliability and validity. Reliability is the extent to which scores for people who have not changed are the same for repeated measurements. A reliable measure minimizes measurement error. Methods of assessing reliability include test-retest reliability (administering a measure twice in a short period to see if the measure yields consistent scores), interrater reliability (assessing whether two raters or observers independently assign similar scores), and internal consistency across items in a composite scale in measuring a trait. Reliability is assessed statistically by computing coefficients that range from .00 to 1.00; higher values indicate greater reliability Validity is the degree to which an instrument measures what it is supposed to measure. Aspects of validity include face validity (the extent to which a measure looks like it is measuring the target construct), content validity (in composite scales, the extent to which an instrument’s content adequately captures the construct), criterion validity (the extent to which scores on a measure are a good reflection of a “gold standard”), and construct validity (the extent to which an instrument adequately measures the targeted construct, as assessed mainly by testing hypotheses). A measure’s validity is not proved or established but rather supported to a greater or lesser extent by evidence Unit 7 Activity 1 Can They or Can t They Variable Manipulation In quantitative research design, the independent variable is manipulated by the researcher to determine if this has an influence on the dependent variable. One important consideration is that all variables are not equal. Some can be manipulated, other cannot. Beside each variable listed write: (C) for can be manipulated or

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture