Turner, J. R., Petrunin, K. F., & Allen, J. (2015). Developing multilevel models for research. In V. Wang (Ed.), Handbook of research on scholarly publishing and research methods. IGI Global, 463-489. doi:10.4018/978-1-4666-7409-7.ch023
Abstract:
In the past, a large number of research e orts concentrated on single-level analysis; however, research- ers who only conduct this level of analysis are nding it harder to justify due to the advancements in statistical software and research techniques. The validation of research ndings comes partially from others replicating existing studies as well as building onto theories. Through replication and valida- tion, the research process becomes cyclical in nature, and each iteration builds upon the next. Each succession of tests sets new boundaries, further veri cation, or falsi cation. For a model to be correctly speci ed, the level of analysis needs to be in congruence with the level of measurement. This chapter provides an overview of multilevel modeling for researchers and provides guides for the development and investigation of these models.
Developing Multilevel Models for Research
| Focal Unit | Level of Analysis | Level of Measurement | ||
| DV | IV | DV | IV | |
| Individual Level | LL | LL | LL | 1) LL |
| 2) HLd | ||||
| Group Level | HL | HL | HL | 1) LLa |
| 2) HL | ||||
| Notes: DV = Dependent Variable. IV = Independent Variable. LL = Lower Level Variable. HL = Higher Level Variable. d = disaggregated. a = aggregated. | ||||