President is misinformed on disability data

Fact Check

To evaluate the president’s statement on Twitter made after the opening of the Ayati Centre, FactCheck consulted the most recent Population and Housing Census (PHC) compiled in 2011 by the Department of Census and Statistics.

The PHC uses a commonly understood definition of disability, as mental or physical impairments that cause daily difficulty in at least one of six areas (see Exhibit 1). The PHC finds the prevalence of disability to be 1.7% in children aged 5-14 years and 1.8% in children aged 5 – 19. The statistic does not assess disability for children under 5 years. Other ways of measuring disability that are used in other jurisdictions may result in a larger assessment of the prevalence than assessed in Sri Lanka, However, this official figure that is available from the government falls well below the 20% claimed by the president.

In searching for sources that might support the president’s claim, we consulted:

The 2016 Demographic and Health Survey (DHS) – this uses a lesser known statistic of Child Disability, defined by UNICEF as functional limitations relative to age specific norms in at least one of ten areas (see Exhibit 1), in children aged 2-5 years. It reports this prevalence to be 22.8%.

The Ayati website – this says, “an estimated 20% of children have some form of mental or physical disability”.

The DHS figure relates only to children aged 2-5 years. Furthermore, the DHS notes that the questions used in the survey are designed to identify children with a higher risk of some form of clinical disability, but are not specific enough to be a diagnostic of clinical disability.

The estimation on the Ayati website reflects observations from approximate prevalence rates that are assessed in other jurisdictions using wider definitions of disability. However, as an estimation for Sri Lanka, this figure could not be verified with reference to available survey data in Sri Lanka or with reference to published research.

It is clear from the context of the speech, the age range of children being referenced, and the direct reference to mental and physical disability, that the president was expressing the statistic within the commonly understood definition of disability that is used by the PHC. But the number he quoted was possibly drawn from either the DHS statistic, which is not assessed for children above 5 years of age, or from the unverified estimation posted on the Ayati website.

In making the statement, the president may have inadvertently depended on sources that were not suitable. On the available statistics, the statement is incorrect. Therefore, we classify it as FALSE.

*FactCheck’s verdict is based on the most recent information that is publicly accessible. As with every fact check, if new information becomes available, FactCheck will revisit the assessment.

Additional Note:

FactCheck has been informed that some academics in Sri Lanka have significant concerns about the suitability of methods used by the Census to assess disability in Sri Lanka. It is beyond the scope of this fact check to comment on the accuracy or validity of the official statistics in the absence of alternative sources of survey data and national estimates.

Exhibit 1: Disability criteria

*Corrections and Updates – 27 February 2020: The previous version of the fact check only included PCS’s reported prevalence for the 5 – 14 age group. The fact check has been updated to include the figures for the 5 – 19 age group as well.

Right of response 

President Media Division’s (PMD) response to the fact check: 

‘‘On February 27, 2020, the FactCheck column classified President Gotabaya Rajapaksa’s statement on Twitter made after the opening of the Ayati centre that 20 per cent of children in Sri Lanka are having a mental or physical disability as false.
You have quoted the Population and Housing Census (PHC) compiled by the Department of Census and Statistics (DCAS) and have stated that according to their findings only about 1.7 per cent in children aged between 5 and 14 are affected with an impairment that causes daily difficulty. As such your newspaper had noted that “this falls well below the 20 per cent claimed by the President”.

However, both the Department of Census and Statistics (DCAS) and Professor in Pediatric Disabilities Samanmali P Sumanasena attached to the Department of Disabilities, Faculty of Medicine, University of Kelaniya disagree with the FactCheck classification. For your reference, statements from both parties are as attached.

The definitions of the indicators given in the two reports namely Population and Housing Census (PHC) and the Demographic and Health Survey (DHS) referred to by FactCheck are different and consist of two reference age groups. This is very important when the indicators are explained noted Additional Director General (Statistics) Anoja Seneviratne at the Department of Census & Statistics. She explained that disability is different from difficulty. The 1.7 per cent of children in the DCAS report is found with only functional difficulties, she observed.

Professor Sumanasena also noted that childhood disabilities range from physical, sensory (visual and hearing), communication, intellectual, learning and social and behavioural. She had further noted that due to a variety of factors the actual number of children with disabilities might be much higher than the estimated 20 per cent. Though FactCheck could not verify with reference to available survey data in Sri Lanka, it is clear that the Ayati estimate is close to the findings of the DHS statistic report. Therefore, this estimate cannot be summarily dismissed as many factors need to be taken into consideration.

We believe that FactCheck is an important column as it aims to give facts and figures to the reader. However, as the above contents show, there are a number of different methodologies taking into account different indices when computing statistics. Therefore, for FactCheck to live up to its reputation as the reliable verification point, and for your esteemed newspaper to give all facts relating to the story, these facts as explained by the relevant professionals must be also given the same prominence in your newspaper as yesterday’s classification.’’

Mohan Samaranayake
Director Genaral – President’s Media Division

Verité Research’s Reply to the President’s Media Division on our fact check

FactCheck appreciates the feedback of the President’s Media Division, the Department of Census and Statistics (DCS) and Professor Sumanasena.

The DCS provides more information on the definitions used by the two surveys referred to by FactCheck. This was already carried in the longer version of the fact check on the web ( https://www.factcheck.lk/claim/gotabaya-rajapaksa-4 – the press version is constrained to be shorter). The DCS parenthetical note that difficulties are different from disabilities might be restating what is known: that it is specific difficulties at specifically reported intensities that DCS classifies as “disability”.

In any case, the DCS is not admitting to having blundered in its methods/estimation of disability and is simply confirming, in all respects, including in reference to the DHS survey, what FactCheck had already noted and used as the basis for its conclusion.

The note from Professor Sumanasena reiterated what FactCheck had already been made aware of and published in the web version. That is: “FactCheck has been informed that some academics in Sri Lanka have significant concerns about the suitability of methods used by the Census to assess disability in Sri Lanka. It is beyond the scope of this fact check to comment on the accuracy or validity of the official statistics in the absence of alternative sources of survey data and national estimates.”

Professor Sumanasena’s note suggests that she disagrees with the method/implementation of the DCS in ascertaining and reporting on disability statistics. It also makes clear what FactCheck had already ascertained: (i) there are no publications in which she or others have contested and corrected the DCS official disability statistics, which are 1.7% for children aged 5-14 (1.8% for children aged 5-19); (ii) the data cited by the professor in support of her position relates to surveys in other countries and a survey of children within a two-year age range in Gampaha. It was after considering all this that FactCheck concluded that there was, as of yet, no credible basis for the public claim that Sri Lanka has a disability prevalence rate of 20% among children.
In light of the above, there is no additional information that has been received to consider updating the verdict of the fact check.

To view the response from the President’s Media Division (PMD), please see:  http://www.dailymirror.lk/…/Right-of-Reply-Presi…/131-184089