Home Study 1 Study 2 Study 3

Do national disability laws matter?

Erika Sanborne, University of Minnesota

disabled people at work

If anything here is not accessible to you, please let me know how I can help you.

Research Questions

1. How does the nature of national disability legislation (social vs. medical model) relate to the educational attainment of disabled women?

2. How does the nature of national disability legislation (social vs. medical model) relate to the subjective well-being of disabled women?

Background

Why don’t national leaders make decisions and policies that truly prioritize the well-being of their people, in a sustainable way? When I was a young, naïve, under-50 year old, I asked that question. Back then, I was expecting that national leaders would tend to act rationally. They do not.

But the question challenges the assumption that national leaders act on rationality. Instead, as Meyer and Rowan (1977) suggest, national actions are often about gaining legitimacy rather than efficiency or well-being. The adoption of national disability laws might appear as an effort to conform to international norms (Finnemore and Sikkink 1998), yet as Hafner-Burton and Tsutsui (2005) identify, these can be mere empty promises, where the actual enforcement of such laws lags significantly behind their adoption.

Critics argue that states might adopt national laws more for international appearances or under external country-level peer pressure, rather than for a real commitment to ever enforce them or follow through with tracking connected progress metrics.

Instead there are these myths in society, rationalized institutional structures, built into rationalized institutional elements (practices, procedures) which lead to rules (Meyer and Rowan 1977:345). And if we do think of the nation-state as a rational actor (Scott 2017:863), nations are going to incorporate selective elements and promote whatever policies imbue them with legitimacy, because survival and legitimacy are sought, not efficiency (Meyer and Rowan 1977) nor ecology, sustainability, accessibility or well-being, by extension.

“Human rights” was institutionalized, for example, in the latter decades of the previous century. Did it lead to a proliferation of international human rights policies, laws and treaties? Yes, it did. What did all of that accomplish? Does humanity have so much human rights now that it’s overflowing with human rights?

Researchers who investigated human rights policy proliferation came up with the “paradox of empty promises” (Hafner-Burton and Tsutsui 2005) to describe what they ultimately discerned was a context of nation-states “ratifying human rights treaties as a matter of window dressing, radically decoupling policy from practice (when some bad human rights practices were even made worse under the cover of these treaties), but the emergent global legitimacy of human rights (led to) independent global civil society effects that (tended to improve) states’ actual human rights practices (in spite of it all)” (1373).

This phenomenon is further underscored by the role of global civil society as highlighted by Keck and Sikkink (1998), who detail how international advocacy networks, including those focused on disability rights, use global norms to press nations into adopting progressive laws. Moreover, Barnett and Finnemore (1999) explore how international organizations can enforce certain practices, which could explain the widespread adoption of national disability legislation that aligns with international standards but may lack effectiveness locally.

In other words, the nation-states signed treaties as window dressing, which sometimes gave concealment for continued violations and sometimes for things to get worse, but despite that, the paradox is that the proliferation of all of this window dressing led, separately, to international civil society acting as agents who pressured governments to make changes for human rights. They could then do this by using the legitimated norms of human rights as codified in those treaties.

It’s a cool paradox. And so even if it is all window dressing, there are civil society agents in the form of disability activists and a few of us scholars eager to use such legitimacy hoping that real change can come next.

This now permanent skepticism extends to the effectiveness of laws due to decoupling, a process in which policies and real practices remain unaligned, particularly in systems with weak oversight. Additionally, the inconsistency between international laws and local, in-country cultural norms and power structures can render these laws ineffective. Overall, the lack of rigorous analysis and the evident cultural and structural barriers contribute to a generally pessimistic view of the impact of national laws and the possibility of change.

Despite some more balanced, healthy skepticism, studies have found the potential of national laws to sometimes drive social change anyway. This optimism is reinforced by the role of laws as strategic tools globally, attracting activists, investors, and funders, thereby legitimizing and accelerating social change efforts. The diffusion of laws across countries in general suggests their probable recognition as strategic mechanisms for change.

Of course, a national law can be present but ineffective as well. The effectiveness of national laws varies significantly depending on their form and the factors influencing implementation. Recursivity plays a role in determining a national law’s impact. This suggests that understanding which types of laws are most effective requires considering their compliance with international standards.

Collinearity in this context describes how national disability laws might align with international standards while not being effective on the ground. This study will seek to investigate whether national disability laws actually do influence quality of life and well-being for the disabled population, or whether they are just more window dressing and responses to country-level peer pressure.

The starting assumption is that many national disability laws will indeed align with international standards such as the Convention on the Rights of Persons with Disabilities (CRPD), but that they might not be effective on the ground. Effectiveness in this case is being operationalized within the outcome of educational attainment. This study will control for (fixed) institutional effects in order to investigate whether national laws that mirror global treaties and norms actually influence local practices and outcomes.

Hypothesis

Hypothesis 1: Disabled women in countries that have a national disability policy with a social model of disability orientation will have higher educational attainment ceteris paribus.

Hypothesis 2: Disabled women in countries that have a national disability policy with a social model of disability orientation will have higher subjective well-being ceteris paribus.

Contributions

This study contributes to understanding the alignment of national disability laws with international standards, particularly the CRPD, focusing on their practical effectiveness and addressing the implementation gap in disability legislation. It connects these discussions with multidimensional well-being and sustainable development, acknowledging both the potential influence and the limitations of national policies that align with global norms such as the CRPD on meaningful well-being outcomes for disabled women.

References

Methods

Data Sources

This study will collect and code national disability laws, as described below. This will lead to the construction of a binary indicator of national disability law type, which will be a key independent variable.

My dependent variable and other variables will come from IPUMS MICS.

Coding National Disability Laws

1. Acquire Data

Starting with the The WORLD Policy Analysis Center Disability Data Download, collect all current national disability laws.

2. Identify Key Features

Medical model features will include language such as that which emphasizes medical treatment, rehabilitation, or citizens’ adjustments to disability. It may also include definitions of terms consistent with a medical model of disability.

Social model features will include language such as that which emphasizes accessibility, removing barriers, ensuring rights or opportunities or freedoms.

India’s national disability law (Rights of Persons with Disabilities Act) has elements of both models. While it includes many things about creating barrier-free environemnts (social model), it has several points of a medical model too. For example:

– Section 58 (Certification of Disabilities) stipulates a process for medical authorities to certify disabilities, emphasizing a medical assessment to establish the presence and extent of disabilities (medical model).

– In Section 56, the act sets guidelines for the medical assessment of disabilities, focusing on diagnosis and classification (medical model)

– Section 57 is about designating medical authorities as certifying authorities, reinforcing the role of medical professionals in the validation and recognition of disabilities (medical model)

More clear and familiar examples of social model coding would be the United States (ADA) and the United Kingdom (EQ) national disability laws, which emphasize removing barriers and allowing access.

3. Create a checklist of features typical of each model

The checklist will allow scoring, one point per feature, so that a law that is predominantly medical model or social model gets coded correctly because it has more features of one than the other.

Possible issues at this point: I may need to create a third, hybrid category for laws that contain features from both lists, as in the India example. Then, after reviewing several policies I can revisit and recode these hybrid laws.

Another possible resolution of issues at this point: I may be able to set aside such “hybrid” laws initially, and then look into the context of their origin. For example, if the law was created in response to something, that could suggest its intent as either a medical model or social model policy.

Although, in the case of the India example, it wsa created in response to the CRPD (2006) itself, which is social model. And yet, when India passed it’s law ten years later, it has medical model context and reasoning embedded.

Key Dependent Variable: Educational Attainment

education: the outcome measure of highest level of school attended by the woman, categorized into four ordered levels: less than primary, primary, secondary, or tertiary/higher/university.

This research will analyze the IPUMS MICS-6 data for all available countries. Please find a Methods discussion: About IPUMS MICS on the home page of this prospectus.

Additional Dependent Variable: Life Evaluation Measure

An additional model will be fitted for the outcome measure of evaluative well-being using the Cantril ladder item, from IPUMS MICS Round 6. See also: Methods discussion: About the Cantril ladder.

My rationale for including an aditional, subjective well-being model is that educational attainment, and subjective well-being, should reveal two aspects of multidimensional well-being, a broader construct that includes them both.

Other Independent Variables

discrimination: binary indicator constructed such that 1=report of having felt personally discriminated against in the 12 months prior to survey administration, for one or more of the six categories surveyed (sexual orientation, age, religion or belief, disability, gender, or any other reason).

wealth: binary indicator constructed from wealth index score to be 1 for respondents with positive wealth index score and 0 for those with negative wealth index score.

marital: binary indicator constructed using marital status, categories were combined so that women were 1=married or partnered, and 0=not married, with the latter including women who reported being divorced, widowed, separated or never married.

healthinsur: binary indicator of whether the woman has health insurance.

unsafe: binary indicator constructed from safe home and safe walk. If the women reported that she feels ‘unsafe’ or ‘very unsafe’ either at home or walking home in her neighborhood after dark, this is 1=unsafe.

Multilevel logistic regression model

melogit education i.lawType i.wealth i.healthinsur i.marital i.discrimination i.unsafe || countryID:, or

logit(πij​)=β0​+β1​LawTypeij​+
β2​Wealthij​+β3​HealthInsurij​+
β4​Maritalij​+β5​Discriminationij​+
β6​Unsafeij​+uj

Where:

  • 𝜋𝑖𝑗πij​ is the probability of achieving a level of education for a woman 𝑖i in country 𝑗j.
  • 𝛽0β0​ is the intercept.
  • 𝛽1,𝛽2,𝛽3,𝛽4,𝛽5,𝛽6β1​,β2​,β3​,β4​,β5​,β6​ are the coefficients for the respective variables LawType, Wealth, Health Insurance, Marital Status, Experienced Discrimination, and Perceived Lack of Safety.
  • 𝑢𝑗uj​ is the random effect for country 𝑗j to account for differences between countries that are not captured by the observed variables.

This model accounts for the hierarchical structure of the data, where individual observations (educational attainment of disabled women) are nested within higher-level units (countries with specific types of national disability legislation). This model can handle variability both within and between countries and control for confounding variables at different levels also.

The same structure of a regression model will be fitted for the binned Cantril ladder outcome.

Limitations

Importantly, this study is not going to be able to make causal claims despite any relationships that may be observed between national disability legislation and educational outcomes.

Also, the binary classification of disability laws into social and medical models risks oversimplifying their complex nature, potentially obscuring the influence of national laws on the education outcome (measurement error).

Lastly, temporal dynamics are a concern in countries where their national disability law was passed recently, because the outcome measure of educational attainment is one that has a temporal lag. The impact of national disability policy on this educational outcome for women would not be immediate.

It is because of this temporal concerns that I have added a second analysis, since the independent variable of interest is about type of national policy, and I want to address this possible unknown time variable regarding the educational attainment outcome.

Timeline

This analysis will be completed Summer 2024.