ABSTRACT
Pedigree problems are typical tasks in school genetics classes. However, they are perceived as difficult and often are not successfully completed. Therefore, the purpose of this study is to determine relevant factors that might have an impact on success in pedigree analysis. Based on previous research, we investigate the influence of the superficial appearance of the pedigree (relating to a well-known misconception), the mode of inheritance, content knowledge, reasoning abilities, the last biology grade, and mental effort. This means, we are simultaneously examining student and task characteristics. For this purpose, we analyse data from N = 135 students, who solved four pedigree problems each, using Generalised Linear Mixed Models (GLMMs). Specifically, we use multilevel logistic regression to determine the influence of the variables in question. The final model shows mixed results: For example, the represented mode of inheritance has no significant influence. The superficial appearance of the pedigree, in contrast, is one of the important predictors of success in identifying the present mode of inheritance. Therefore, our results imply that students make decisions based on misconceptions when analysing pedigree problems. This and all other results are discussed in order to infer implications for teaching and learning pedigree analysis.
Acknowledgments
The authors would like to thank Dr Yvonne Lettmann for her advice on the design and realisation of the study and Prof. Dr Christian Johannes for his professional advice and discussion on pedigree analysis. Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the authors and do not necessarily reflect the views of the funders.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Ethics statement
In accordance with ethical and legal guidelines, all participants were informed in advance about the aim and procedure of the study and participated voluntarily. The results of the individual courses were made available to the teachers in anonymous and aggregated form. Inference to individual students was precluded. Only necessary data was collected and handled in accordance with local data protection laws.
Notes
1 A small note on our procedure: We would have preferred to investigate the influence of the task level predictors by comparing non-nested models that include both pedigrees and students as random effects. In this case, we could have determined estimates for the fixed effects as well as changes in the variance of the random effects after including our predictors as fixed effects. However, probably because of the small number of pedigrees, we observed singular fit (zero variance in the random effect of pedigrees) when including the predictor pedigree appearance into a non-nested random effects model. A random effect with zero variance should not affect the other model parameters and a reduced variance in the random effect seems reasonable after including a significant predictor. Nevertheless, singular fit models should not be used, and we felt it was advisable to switch to fixed effect models (Oberpriller et al., Citation2022).