**5. Acknowledgments**

190 Learning Disabilities

without LD), statistically significant differences are observed, both for the writing tasks as for the comparison-contrast texts. In comparison-contrast texts, significant differences were observed regarding the variables of information generation, organization of ideas, structure or reader-based quality. Examples include the organization of ideas in relational coherence, reader-based evaluation concerning structure, and reader-based evaluation for order and quality. Also, in the writing task statistically significant differences were found regarding idea generation and total productivity, in reader-based evaluation for order, structure and quality. As far as teachers are concerned, it is interesting to observe that they behave differently based on the students they have, as regards motivation, general or natural teaching approach and self-efficacy. The tests of intersubject effects indicated statistical significance for the opinion variables, in the subcomponents of motivation; approach,

When trying to extract some pattern that would help understand the relationship between what teachers actually do when they teach writing and students' achievement, a very interesting observation was made. A relationship was noted underlining the fact that some total measures of the PRAES predicted some totals of the writing product. Other variables concerning the teachers that helped to predict some measures in writing achievement were the procedures used, the role assigned to the family, the materials employed, the natural teaching approach, personal self-efficacy, the formal approach, teacher training, motivation, the practice and the use of plans and revisions. Of these, those that most significantly predict writing variables are: the family, the activities carried out in the classroom along with the procedures and the materials used, as well as the use the formal theoretical

When the type of students (ADHD, with and without LD) was taken as an independent or a predicted variable from the set of total measures of the writing product we found variables included in the model related to writing: writing a draft, reader-based evaluation, order and structure and other aspects of total coherence. We also noted variables related to the PRAES: practice in the classroom in the procedure subcomponent; opinion and the motivation subcomponent. The remaining variables were excluded from the model, as they did not reach statistical significance. When we attempted to predict writing achievement through total productivity in the 'number of words' parameter, a regression model or equation was obtained with the following PRAES predicting variables: opinion, the family subcomponent, and the formal approach. No other variable was found to be significant. These factors are, at least indirectly, based on the type of teachers' practice in the area of Spanish language. There were no significant differences in relation to the theoretical concepts and strategies applied in the teaching of writing in the different school years from Infant to Primary. There were, nevertheless, differences based on age, which is an indicator of some type of adjustment according to the students' stage of development. According to the results obtained, all teachers seem to act in a homogenous way as far as the theoretical conceptions that direct their behavior are concerned - or in other respects, which indicates poor selfregulation of their practice. In addition, it appears that their practice of teaching writing does not substantially differ between students with and without learning difficulties and/or

It is clear that this study presents evident limitations concerning the use of PRAES, and it should be complemented with direct observation of the teachers' behavior as well as direct

natural learning; and general self-efficacy.

approach.

low levels of achievement.

During this research study, we received competitive funds from the Spanish Ministry of Education, Science and Innovation (MICINN) (EDU2010-19250 / EDUC) for 2010-2013, and Excellence Research Group funds from the Junta de Castilla y León (GR259), with FEDER funds from the European Union for 2009-2010-2011 (BOCyL 27 on April 2009). Both were awarded to the Director/Main Researcher (J. N. García). We are very grateful to Victoria Rosa Sturley for her help with the English version.

*Correspondence* should be addressed to Departamento de Psicología, Sociología y Filosofía. Área Psicología Evolutiva y de la Educación. Campus de Vegazana s/n, 24071 – León, Spain. Phone: +34-987291041 (university); +34-987222118 (home); +34-652817871 (cell); Fax: +34-987291035 (university). E-mail: jn.garcia@unileon.es.
