Comparability — Armenia [top]

The 2011 sample identified respondents who reported speaking English as their mother tongue or as their second language.

Comparability — Belarus [top]

The 2009 sample identified respondents who reported speaking English as one of their fluent languages. Respondents could report up to four fluent languages.

Comparability — Bolivia [top]

The 2012 asks about languages that the person speaks and allows for up to five responses, including persons who speak English.

Comparability — Burkina Faso [top]

This sample identified respondents who reported speaking English as their main language.

Comparability — Cambodia [top]

The 2008 and 2019 samples identify individuals who are literate in a language other than Khmer. Literacy is defined as the ability to both read and write a simple message with understanding in the language. Those who identified English as a language they are literate in, or persons who indicated English as their mother tongue (regardless of whether or not they also listed English literacy), are coded as "yes."

Other samples for Cambodia are not able to be included in SPEAKENG due to confidentiality issues.

Comparability — Cameroon [top]

The 2005 sample asked whether respondents could speak English. The level of English required for an affirmative response is not defined, although whether the person could read or write English is available in the unharmonized source variables. The universe is limited to household residents age 12+.

Comparability — Canada [top]

The questionnaires asked whether respondents could speak English well enough to carry on a conversation. All samples report French and bilingual French-English in the unharmonized source variables.

Comparability — Ethiopia [top]

This variable is only available for the 1994 sample. The data identify respondents who reported speaking English as their mother tongue or as their second language. The 2007 census asked for mother tongue only and is not included in this variable. See the unharmonized source variable for this information.

Comparability — Ghana [top]

The source variables for the Ghanaian samples report language of literacy. The universe changes from persons age 15 or more in the 2000 sample to persons age 11 or more in 2010.

The 2000 sample unharmonized source variable identifies Ghanaian and bilingual English-Ghanaian responses. The 2010 sample unharmonized source variable distinguishes between literacy in English; English and Ghanaian; English and French; and English, French and Ghanaian.

Comparability — Guinea [top]

The 1996 sample reports whether individuals were literate in English or English in combination with some other language. All responses that included English are coded "yes." The combinations are available in the unharmonized source variable.

Comparability — Hungary [top]

The 1990 sample asks about mother tongue, in addition to the first, second, and third language spoken other than mother tongue.

Comparability — Ireland [top]

The 2011 and 2016 samples question asked respondents to specify a level of English, ranging from "very well" to "not well." Respondents reporting any level of English were categorized as "yes" in the integrated variable. There are a large number of cases with missing values. See the unharmonized source variable for more detail.

Comparability — Kyrgyzstan [top]

These samples identify individuals who speak English as a first, second, or (in 2009) third language.

Comparability — Malawi [top]

All three samples report either yes or no to whether the respondent is literate (able to read and write a short sentence) in English. They are not, however, consistent in their universe statements. In 1987 and 1998 the question was asked of those 5 and older and in 2008 the question was asked of residents who were 3 and older.

Comparability — Malaysia [top]

Both samples report detailed combinations and language skill levels in the unharmonized source variables. Any response that includes English is coded as "yes."

Comparability — Mauritius [top]

The 2011 sample asks whether the respondent is literate in English.

Comparability — Morocco [top]

The 2014 sample asks about the languages in which the person is literate and allows for up to three responses, including literacy in English.

Comparability — Papua New Guinea [top]

The data identify persons who are literate, or those who can read or write, in the English language. The 1990-2000 are comparable across years aside from slight universe changes.

In 1980 census asked about the language spoken at home and at the market. See LANGPG1 and LANGPG2 for these data.

Comparability — Paraguay [top]

This variable is available only for the 2002 sample. The data report whether or not the respondent spoke English as their primary language or as their second, third, fourth or fifth language. The 1962 and 1992 samples include variables describing the main language spoken in the household. See the unharmonized source variables for these data.

Comparability — Philippines [top]

The person must be able to communicate in English, even if with a very limited vocabulary.

Comparability — Puerto Rico [top]

The 1970 sample indicates whether or not the respondent spoke English. The 1980-2020 samples report how well the respondent claimed he or she could speak English. The 2000-2020 samples provide additional detail, reporting whether the respondent spoke only English. All respondents who indicated they spoke any English are coded as speaking English in SPEAKENG, but the full detail is available in the unharmonized source variables.

Comparability — Russia [top]

Both 2002 and 2010 samples indicate persons who speak English fluently as a first, second, or third non-Russian language.

Comparability — Rwanda [top]

The 1991 and 2002 samples identify languages spoken by the respondent, allowing for multiple responses (which include English). The 2012 sample identified persons who are literate, or those who can read or write, in the English language.

Comparability — Sierra Leone [top]

The 2015 sample identifies persons who are literate, who can read and write, in the English language.

The 2004 and 2015 samples recorded the primary and secondary spoken language, but these responses may not include every person who is capable of speaking English. This information is not harmonized into SPEAKENG, but it is available as unharmonized source variables for each sample.

Comparability — South Africa [top]

This variable is available for the 1996 and 2011 samples. The data identify persons who spoke English as their first or second language. The 2001 census asked respondents to indicate their first language only. See the unharmonized source variables for these data.

Comparability — Switzerland [top]

The 2011 sample identified respondents who reported speaking English as one of their fluent languages. Respondents could report up to three fluent languages.

Comparability — Tanzania [top]

The data identified persons who are literate, or those who can read or write, in the English language.

Comparability — Togo [top]

The census asked whether respondents could read, write, or speak English.

Comparability — Ukraine [top]

This variable indicates whether respondents reported English as a second language of fluency. A small number (less than 300) of respondents reported English as their primary language, but these persons are not identified for confidentiality reasons.

Comparability — United States [top]

The census question asked respondents to specify the ability to speak English, ranging from "not at all" to "very well." Respondents reporting any level of English were categorized as "yes" in the integrated variable. See the unharmonized source variables for more detail.

Comparability — Uruguay [top]

The universe for Uruguay is restricted to persons in the labor force.

Comparability — Zambia [top]

For the 1990 and 2000 samples, the data report persons that identified English as their primary or secondary language. In 2010, respondents were asked about their main spoken language only.

The universe changes from persons age 3 or more in the 1990 and 2000 samples to present members and visitors in the 2010 sample. The 1990 and 2000 enumeration instructions indicate that this question should have been asked only of present household members and visitors, but ability to speak English is reported for almost all absent household members; they are included in the universe.