Observers Say Burma’s Census ‘Successful’—Except in Arakan State

By Yen Saning 18 August 2014

RANGOON — Burma’s first nationwide census in three decades was conducted successfully and in line with international standards across the country, with the exception of Arakan State, according to a team of independent observers.

A team of 46 national and international monitors witnessed how the census was conducted between March 30 and April 10 in all 15 of Burma’s states and divisions. The Independent Observation Mission witnessed a total of 2,193 interviews in 901 different enumeration areas in 121 townships, according to a statement dated Aug. 14.

“The Mission described the Myanmar Census as successful on the whole and in line with international standards, except in Rakhine [Arakan], where almost all communities that wanted to self-identify as ‘Rohingya’ (who the Government call Bengali) were not counted,” said the statement distributed by the UN Information Center in Rangoon.

“At the time of the observation, it was noted that some parts of Kachin State, controlled by the Kachin Independence Organisation, were not enumerated.”

Rohingya are not included in the list of 135 official ethnic groups in Burma. In a last minute change of policy, the government announced that the name many in the Muslim minority in Arakan State use to refer to themselves be excluded from the UN-backed census.

“By not allowing these specific subpopulations to self-identify and be counted, the census in these areas fell short of international standards,” the mission’s report said.
“If the missing populations are not included, based on a proper count or estimation, the resulting undercount will have a negative impact on the census results at the State and Region level and the national level.”
Outside of Arakan State, the observers said the census enumeration process was well organized and the enumerators, trainers and supervisors were well trained. Census awareness raising activities were effective and the public were “generally positive about the census.”

However, they did note some shortcomings in the census process based on their observations. The observers said census enumerators “seldom explained the census,” failed to “explain the concept of confidentiality of response,” and did not ask specifically about the night of March 29, the chosen night that the so-called “de facto” census was supposed to record.

“[A] large portion (31 percent) of the enumerators did not systematically refer to the census night to determine who was in the household on that reference night, which is a core concept of a de facto census,” the monitors’ 124-page-long report said.

The observers’ report said that sometimes enumerators also filled out the forms using inferences rather than directly asking the respondents.

“Questions on religion, ethnicity, education and household characteristics and assets were sometimes inferred or directed from what the enumerator could observe (for instance, after asking the ethnicity of the head of household and spouse, the enumerator filled in the ethnicity of their biological children without asking),” it said.

“Additionally and even though most respondents could self-identify their ethnicity, the observers noticed that most enumerators only recorded the main ethnicity rather than the sub-ethnicities.”
Salai Isaac Khen, a member of National Technical Advisory Board for the census, said he agreed with the findings of the observation team, which chimed with problems he encountered while monitoring the census in Chin State.

“[Enumerators] did not emphasize questions on disability according to the guidelines. They also did not ask specific questions on ethnicity, and just record it with their own guess as enumerators are locals,” he told The Irrawaddy.

Salai Isaac Khen, who is also coordinator of the Chin national supporting committee on the census, said these issues meant that while gender and population figures in the census would likely be accurate, ethnicity and religion results might not be acceptable.

The question of ethnicity in the census has been highly controversial, drawing 80 complaints nationwide, he said, and the census commission has decided to withhold the results on ethnicity from the main census report, due to be published next year.

Ethnicity data will be released in 2016—after the national elections—following a review of complaints by a team consisting of parliamentarians from the ethnic affairs committee, ethnic representatives, academics and experts.

“We are trying to avoid the use of ethnic data [from the census] in the coming election’s voter list,” Salai Isaac Khen said.