At Odds: School Achievement -- Bad Data Must Be Challenged
Notice bibliographique
Résumé
No party line is required to challenge bad data, badly collected, badly analyzed, and badly reported, Ms. Robertson retorts. SINCE HE is so eager to draw attention to my organizational affiliations, let me remind readers that Gilles Fournier is the coordinator of the national testing program that he defends so vigorously. As my critique of the School Achievement Indicators Program (SAIP) had to do with its substance, rather than its provenance, I saw no reason to mention Fournier's name in my column. This may change, however, since Fournier has served up a number of quite astonishing quotable quotes that I may not be able to resist citing in future commentary on the Council of Ministers of Education Canada (CMEC) and SAIP. For example, he claims that, since I find numerous faults with the national testing program, I am obviously opposed to student evaluation of all kinds. He then makes the equally silly statement that the act of administering tests that teachers have neither designed nor marked constitutes training teachers to assess their pupils properly. Somehow I doubt that Fournier's credibility in the evaluation field has been enhanced by these remarks. Fournier defends his program by pointing to results that have found that differences in scores are associated with gender and linguistic differences. I can only hope that these results did not surprise him or his office, any more than did the finding that 16-year-olds know more than 13-year-olds. Yet even this spirited defense avoids the pretense that SAIP provides anyone with the slightest idea of how to alter persistent achievement gaps. Indeed, I note that Fournier evades entirely the matter of how six SAIP assessments have been used to inform policy and improve practice, unless designing curriculum around questions found on standardized tests constitutes policy making. With respect to Fournier's objections to my depiction of the expectations-setting process, I feel obligated to warn my critic that he does his case no favor by providing readers with more details about a process so bereft of validity and reliability. I believe others will dispute his claim that this process is consistent with the Modified Angoff, which is used to determine cutoff scores - not the percentages of students who should achieve at predefined levels. Even the 1997 review of SAIP, commissioned by CMEC, recommended changes to enhance the adequacy of the expectations-setting or standards-setting process to deal with bias and to address the validity and reliability of the expectations. Robert Crocker pointed out that, in the absence of confidence intervals associated with these expectations, there is no way to determine whether the reported differences between expected and achieved results are statistically significant. …
Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.
Comment cette classification a été obtenuedéplier
Prédiction machine sur la base complète
Imitation des enseignantsNi prévalence calibrée, ni vérité terrain. Validation humaine à venir. Le volet Gemma est une étiquette directe du modèle pour chaque travail de la base, lue sur la notice réduite au titre. Le volet Codex est un classifieur appris des 10 348 étiquettes directes de Codex et calibré sur les taux pondérés de l'échantillon; les champs sans appui suffisant ne portent aucun appel Codex. Le mode candidate est l'union des deux volets; le consensus est leur intersection. Ces sorties portent le statut machine_predicted_unvalidated et ne sont pas des étiquettes humaines.
Scores du classifieur distillé par catégorie (deux têtes)
| Catégorie | Codex | Gemma |
|---|---|---|
| Métarecherche | 0,219 | 0,501 |
| Méta-épidémiologie (sens strict) | 0,001 | 0,001 |
| Méta-épidémiologie (sens large) | 0,004 | 0,002 |
| Bibliométrie | 0,007 | 0,008 |
| Études des sciences et des technologies | 0,012 | 0,055 |
| Communication savante | 0,032 | 0,045 |
| Science ouverte | 0,008 | 0,016 |
| Intégrité de la recherche | 0,019 | 0,066 |
| Charge utile insuffisante (le modèle a refusé de juger) | 0,006 | 0,005 |
Scores machine (provisoires)
Les deux têtes enseignantes du modèle étudiant, lues sur ce travail. Un score ordonne la base pour la relecture; il n'affirme jamais une catégorie, et le statut de validation accompagne chaque rangée tel quel.
Scores de référence d'un modèle non mature (critères de maturité non atteints, 7 itérations). Un score ordonne; il n'affirme jamais une catégorie.
score_only:v0-immature-baseline · tel quel depuis la passe de notation : score_only signifie que le nombre peut ordonner les travaux, et qu'aucune étiquette de catégorie n'en découleClassification
machine, non validéePrédiction automatique; un appel candidat d’une seule source (Gemma direct ou Codex distillé), pas un consensus.
Le détail, modèle par modèle et score par score, se trouve en fin de page sous « Comment cette classification a été obtenue ».