Conducting time series research
The time series design is a research design in which the measurement of the same variables at different points in time are taken, usually with a view to studying social trends and thus formulation of an intervention mechanisms. That is, a research design in which periodic measurements are taken on a defined group of individuals, both after and before implementation of an intervention (Brockwell, 2002). At times they are referred to as trend designs and different from “one shot” cross-sectional designs in which measurement are only taken once.
The time series design analysis is a statistical methodology suitable for an important category of longitudinal research designs. The designs typically involve research units or single subjects that measured repeatedly at some regular intervals over a big number of observations (Colby & Velicer, 2001). The time series analysis helps inunderstanding the underlying naturalistic process, the trend of change over time, or the evaluationof the effects of either unplanned or planned intervention.
This paper determines a successful behavioral intervention design for Kaya in the reducing her off-task behavior, physical aggression, and verbal aggression.
Three designs of time series research
For the purpose of this research, we look into the general model that allows for three design variations:
This is a time series design described by Huitema & McKean (2000), in whichthere is use of the observed trends or cycles in a series in inferring the nature of a latent causal mechanism.
This second time series research design attempts to deduce a causal relationship existing between two series from their covariance. The validity of the correlational inferences leans highly on the theory (Fan & Cook, 2003). In a case where theory depicts a single causal effect that operates at discrete lags, just like in these natural examples, correlational research design backs up unambiguous causal interpretations. However when there is lack of theoretical specification, correlational design gives no strong causal inferences.
This time series design deduces the latent causal effect of a temporally discrete treatment or intervention from interruptions or discontinuities in a time series (Menard, 2002). This design involves selection of groups, upon which testing of a variable is carried out without any process of random pre-selection.
Download Full Sample Paper On Conducting Time Series Research HERE Or Order original paper on the same topic or any other topic by clicking on the Order Now button. Order Unique Answer Now