.png){kind=icon}
返回
How to Make Recommendations for Educational Practice from Correlational Data Using Structural Equation Models
.png){kind=icon}
PRE
AI
DOI：10.1007/s10648-023-09770-0.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
In this commentary, we outline a five-phase process by which recommendations for educational practice can be distilled from correlational data using structural equation modeling (SEM). First, meta-theoretical beliefs associated with latent variables-that mental attributes cause behavior and can therefore be measured indirectly by observing multiple indicators of that behavior-must be adopted and made explicit. Next, an SEM must be formulated with relevant pathways and covariates that exhaustively represent our theoretical knowledge and assumptions about the structure of the psychological phenomena being studied. Third, model-data-fit indices and estimated parameters associated with the SEM should be carefully interpreted. Fourth, the model should be replicated across educational contexts, and any necessary changes should be incorporated into the relevant psychological theory. Fifth, the results of multiple studies can then be interpreted together with other sources of evidence as a basis for communicating our current theoretical understanding and caveats to practitioners. We also point out that educational recommendations should likely never be entirely prescriptive, and instead lie on a continuum of specificity based on the strength of the evidence.
Keyword:
Structural equation modeling
Recommendations for practice
Research design
Causation
Correlational Data
期刊
IF:
8.8
论文数:
1.5K
被引数:
1.3W
机构
引用论文
.png){kind=icon}
Exploring Factor Model Parameters across Continuous Variables with Local Structural Equation Models
PREAI
分享收藏
查看更多内容