School gender discrimination

Do you have questions or comments about this model? Ask them here! (You'll first need to log in.)

WHAT IS IT?

This model demonstrates that discrimination by teachers can result in better school notes for girls, while at the same time assuring that boys achieve higher scores at state exams.

It also takes into account the "family suport", and thus can also show that children from "suportive" families will achieve better results than those which are "neutral" or "discouraging"

MOTIVATION

In Austria, the school system sorts the children after only four years. In these four years, they are tutored by a single, mostly female, teacher and the sorting criterion are the school notes. There is no central exam at the end of the four years elementary school and to the best of my knowledge there are no mechanisms for assessing the quality of teachers' work and sanctioning the discrimination.

Furthermore, the observed situation is:

1) Large portion (TODO: add citation) of the children receives additional paid-for tutoring. 2) Children from socially weaker families have far lower chance of an academic career in Austria than in other OECD countries (TODO: add citation). 3) Boys score much better than girls on non-partial exams such as PISA

The motivation behind this model was to show that the observed situation in Austria can be replicated in a simple model assumes that teachers discriminate against "defiant" kids and that the kids from the better-off families can compensate against the discrimination because they receive additional support from their families.

Assumptions

This model is based on real-world observations that the boys are discriminated by prevalently female and order-loving teachers because of their tendency to be less system-conform than the girls.

For more information, see e.g. http://ideas.time.com/2013/02/06/do-teachers-really-discriminate-against-boys/

Other model assumptions are the following:

1) There is no gender difference in learning aptitudes (modelled by IQ parameter) 2) There is also no gender difference between reactions to discrimination, but some kids (from "supportive" environment) react by learning more, e.g. because they are helped by parents/tutors. Likewise, some kids react by learning even less than they would othervise because they live in "discouraging" environment, e.g. because their parents are less educated, can't afford the tutoring fees and don't believe that the education is very important. 3) Teachers love order and therefore discriminate against defiant kids and positively discriminate the docile ones. 4) There is no correlation between defiant/docile behavior and supportive/discouraging environment. 5) Supportivesupportive/discouraging environment plays a role for all kids with below-average school success. 6) In addition to "family support", the discriminated children also "learn more" on their own and thus at least partially compensate the discrimination. 7) The children with higher IQ can compensate somewhat better than those with lower IQ (IQ is a synonymous for learning aptitude in this model)

Model predictions

If the model is run with "balanced" parameters, the net results are:

1) Chance of girls getting to secondary-level school are somewhat higher than those of the boys. Same for the kids from supportive families. 2) The docile kids from supportive families can make it to a secondary level school in spite of being less intelligent. 3) Consequently, the small portion of the defiant children which makes it to the secondary school is extremly intelligent and even more knowledgable, whereas the groups of "girls" and "from supportive families" are on average much less intelligent. 4) Score of boys at non-discriminatory tests are better than those of the girls.

Recommended balanced settings are:

• significant portion of boys (30-35%) are defiant but only few girls (5%); the other way around for docile;
• discrimination (15-20%) is somewhat higher than support- and self-support modifier (10-15%).

HOW IT WORKS

The model assumption is that both the girls and the boys exhibit the same school aptitude, but differ in percentage of "defiant" and "docile" individuals.

The teachers are order-loving and therefore tendentially assign better then deserved notes to "docile" pupils, and lower than deserved notes to "defiant" ones. Pupils who fit in neither of the categories receive just grades.

Children from "supportive" families will receive support from their families if their scores are below-average, independently of the reason for lower score. Those from "discouraging" families will receive further penalty.

Finally, the defiant pupils can also learn more and thus achieve somewhat higher scores and knowledge than they would otherwise.

The school system has two levels. Admission to the secondary school is selective according to school notes. Thus, the low-performers in the primary school are eliminated from the secondary level school.

At the end of the secondary level, the pupils absolve a state exam which measures the "knowledge" without the teachers’ discriminatory bias. Without bias, the knowledge would be equal to IQ, but the kids from supportive families and those which are discriminated against by teachers will receive a family-support and self-support bonus.

HOW TO USE IT

1) Set up the model by choosing the values on various sliders at the left-hand side of the GUI 2) Press "setup" 3) Press "Run" 4) Analyze the results

THINGS TO TRY

What happens if the discrimination is strong and self-support and family-support weak?

How does the situation change if the self-support or the family support are very strong? What is the net difference between effects of family- and self- support?

Which parameters give the best match for observed situation in country X? Do these settings they make sense, concerning the state and organization of the countries' educational system? (If not - why?)

EXTENDING THE MODEL

MINT vs. social subjects

This model could be easily extended to incorporate the difference between MINT subjects, where discrimination is more difficult and the social subjects where the discrimination is much easier becasue the grading is inherently more subjective.

(_Grading of a language essay is inherently more subjective than grading of a math exam. The difference may be less pronounced in other cases, but it's always easier to add or subtract the points based on the "i (don't) like the way this answer is written" in non-exact subjects. Unless the exams are strictly multiple-choice, which is not the case in Austria.)

The anticipated prediction of such model is that boys tendentially score better in MINT subjects and lower in less exact ones than girls. Consequently more boys will attend the mint-oriented higher schools, while a majority of the girls ends up in the social studies. This is also in-line with observed behavior, which is quite surprising since the model does not come with a-priory assumption that boys are more interested in MINT than girls or that their families have different expectations!

Luck, time, family expectations...

The model is almost fully deterministic. In real life, the "luck" often plays a role. Adding "luck" factor to the state exams may be a good idea, because these exams are one-off event and simply having a "bad day" could influence the results.

Adding stochastics to school notes would be more difficult - in order to do this, the model would have to introduce a series of tests in each of the schools and then calculate the averages. At a same time, such model would be more realistic in a sense that the reactions of the pupils and their parents to bad notes and teachers discrimination could be established gradually, like in a real life.

NETLOGO FEATURES

In this model, ticks are used only to assure the plots are updated. It also demonstrates the use of histograms and gauss random distribution (for IQ).

RELATED MODELS

TODO

HOW TO CITE

If you mention this model in a publication, please include these citations for the model itself and for the NetLogo software:

• Havlik, D. (2015). School gender discrimination. http://modelingcommons.org/browse/one_model/4242
• Wilensky, U. (1999). NetLogo. http://ccl.northwestern.edu/netlogo/. Center for Connected Learning and Computer-Based Modeling, Northwestern University, Evanston, IL.

COPYRIGHT AND LICENSE

Copyright 2015 Denis Havlik

This work is licensed under the Creative Commons Attribution-NonCommercial-ShareAlike 3.0 License. To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-sa/3.0/ or send a letter to Creative Commons, 559 Nathan Abbott Way, Stanford, California 94305, USA.

To inquire about commercial licenses, please contact Denis Havlik at denis@havlik.org.

CREDITS AND REFERENCES

This model is indirect consequence of the voluntary engagement of my wife and to a lesser degree my own voluntary engagement in the Wiener Lerntafel (http://www.lerntafel.at/home.html). This voluntary organization helps socially disadvantaged children to achieve better school results by providing free tutoring. My experience with these children has convinced me that the Austrian school system is highly unjust and needs to be repaired.
My interest in the “gender discrimination” aspect is a consequence of the experiences with the discrepancies between knowledge and school notes of my own children.

There are 3 versions of this model.

Attached files

This model does not have any ancestors.

This model does not have any descendants.