Ethnic Evidence An Approach to Finding Our People in the Data

Increasing the visibility and value of New Zealand’s diversity

The Ethnic Evidence report is the Ministry’s first major attempt to develop baseline indicators for Ethnic Communities. This required us to overcome some challenges regarding how the system captures and reports data relating to Ethnic Communities, which makes many of our people invisible in the data. This paper sets out how we went about finding Ethnic Communities in the data.

Ethnic Communities consist of the following groups: Asian, MELAA (Middle Eastern, Latin American and African), and Continental European. While it is common for government agencies to report data that includes an Asian grouping, it is rare that MELAA data is reported separately due to low numbers. Along with this, not all African people are included in the official government ethnicity grouping of MELAA. Where possible, we included missing African people. We have called this group MELAA+ in the report. Ethnic Communities and Continental European are also not currently recognised as an official government ethnicity grouping. Data relating to the two largest Asian ethnic groups in New Zealand, Chinese and Indian, are also rarely reported or made available separately.

We worked with government agencies to breakdown the over 120 measures developed for the Ethnic Evidence Report by as many sub-groups of Ethnic Communities as possible. A key technique was the use of combined survey data from multiple survey years, which increased the sample size for smaller groups such as MELAA. This allowed us to obtain information about Chinese, Indian, Other Asian, Continental European, and MELAA peoples, which would not have
otherwise been possible from surveys such as the New Zealand Crime and Victims Survey (NZCVS). Similarly, combining New Zealand Health Survey (NZHS) data allowed us to obtain insights about African, Continental European, Latin American and Middle Eastern people.

One of the tradeoffs with the approach of combining data from multiple survey years was that it gave a single result for the whole period of interest. It did not allow us to assess trends over time for these smaller groups. However, this approach still greatly increases the visibility of these groups in the data and means that in the future, it will be possible to assess trends for these smaller groups now that we have these baseline measures.

Finally, in order to gain insights into business ownership, business profitability and other characteristics, we worked with the Ministry of Business, Innovation and Employment (MBIE) to develop a dataset* from the IDI/LBD which linked businesses to their owners’ data and allowed us to identify the ethnicity of approximately 40% of business owners. Given the low linkage rate, we have treated the results with caution, but they do represent novel findings that are highly relevant in the current policy environment.

[*The dataset was created in the IDI/LBD. The IDI (Integrated Data Infrastructure) is a linked collection of anonymised datasets about people and households. The LBD (Longitudinal Business Database) is the equivalent of the IDI but for data about businesses. The IDI and LBD are linked via tax data.]

MELAA

There is limited reporting about the MELAA population. We worked with the various agencies who collect data from key surveys to obtain MELAA data, ideally using our MELAA+ grouping. Using the Ministry’s MELAA+ grouping increases the value of MELAA data by increasing sample sizes as an additional 50,000 African people are included in this grouping.

For the first time, we were able to report MELAA findings in terms of outcomes relating to experience of crime, health and belonging.

The most interesting findings from the combined survey data approach were a statistically meaningful difference in rates of hazardous drinking, illicit drug use and daily smoking for MELAA relative to the general population (MELAA rates were lower in each case).

We obtained custom annual-level MELAA+ grouped data from the Household Labour Force Survey (HLFS) and Household Economic Survey (HES). Here, we found that MELAA employment, labour force participation rates and median household equivalised disposable income slightly increased/improved when comparing 2023 figures
with the baseline. In addition, annual-level MELAA data from the education and public sector showed small positive changes in MELAA tertiary participation.

We also reported baseline-only data, such as 2018 Census data, which showed MELAA homeownership rates were similar to Māori and Pacific Peoples rates and much lower than European rates. The 2018 Ministry of Housing and Urban Development (HUD) data showed that MELAA people were slightly more likely to be homeless than people from other ethnic groups.

Concerningly, the 2021 What About Me survey data showed that MELAA youth were less likely to have good or excellent wellbeing than the total youth population and had a greater prevalence of severe mental illness than any other ethnic group (treat finding with caution due to sample size issues).

Business data from the IDI/LBD provided some interesting insights into MELAA-owned businesses. There was a small increase in the proportion of MELAA-owned businesses from 2018 to 2022. In general, MELAA-owned businesses were typically smaller than average. Sales and profit characteristics indicated that they are concentrated in industries with lower net profit margins. MELAA-owned businesses have relatively high gross profit margins which indicates that they outperform other business in terms of efficiency.

We were able to obtain New Zealand Health Survey and education sector data for Middle Eastern, African and Latin American subgroups. While, in many cases the small sample sizes meant that we could not draw strong conclusions, future iterations of the Ethnic Evidence report will build on this initial data.

There was no information about the Continental European (CE) population of New Zealand to inform the Ethnic Evidence report as it is not a government ethnicity grouping. We worked with agencies to create this grouping for the first time.

To overcome the constraints of the small CE population, we used the combined survey data approach. This limited the conclusions we could make about changes over time but provided a first view of the CE population.

The most reliable finding from this data was that the CE population has a rate of hazardous drinking that is lower than those of the total population. Other combined survey data outcomes from the NZHS included the proportion of adults with unmet need to see a general practitioner.

We obtained customised New Zealand General Social Survey (NZGSS) data for 2021 on the experience of discrimination in the past 12 months. We found that a lower proportion of CE adults experience discrimination than other Ethnic Community sub-groups and that this difference was statistically meaningful. Data from multiple surveys was not available, so we were not able to assess any trends over time.

As with the MELAA grouping, we used HLFS data to explore changes in employment rates for the CE ethnic grouping. We found that, relative to 2018, CE employment rates have dropped slightly by 2023, after rising during the COVID-19 pandemic.

Asian

People of Asian ethnicity make up the majority of the population covered by the Ministry for Ethnic Communities. As a relatively large ethnic group, it has become common practice for government agencies to report survey results and other key data with an Asian ethnic grouping in recent years. This made it less challenging for us to obtain data about people of Asian ethnicity to inform the Ethnic Evidence report.

While obtaining annual data for the Asian population was relatively easy, we also used the combined survey data approach so that we could compare results with the MELAA population in some cases. Where it was possible, we also obtained additional customised data to provide us with insights into specific Asian populations, usually Chinese and Indian. As we have done for the MELAA+ group, we also obtained novel data from the IDI/LBD relating to Asian business ownership.

Since 2018, we observed positive trends for a number of indicators such as employment, education, and income. There was also evidence of a reduction in the ethnic and gender pay gap and increases in individual net worth.

In health, we observed generally positive trends for Asian people, apart from areas such as high levels of psychological distress and a reduction in physical activity.

We also reported baseline only data, such as 2018 Census data which showed Asian home ownership rates were lower than European rates, and HES data which showed that a higher proportion of Asian households spend more than 50% of their disposable income on housing costs than any other group.

MBIE-sourced business data provided some interesting insights into Asian-owned businesses. Between 2018 and 2022, Asian-owned businesses experienced increases in both average sales revenue and average profit. In general, Asian-owned businesses typically had less than 20 employees and were concentrated in Auckland. Asian-owned businesses have sales and profit characteristics that indicate that they are concentrated in industries with lower net profit margins.

We were able to obtain combined survey data and annual-level data for Chinese and Indian people for some indicators, reflecting their large share of the total Ethnic Communities population. Indian and Chinese employment and unemployment rates improved slightly over the 2018-23 period and largely followed patterns seen in other ethnic
groups regarding the impact of the COVID-19 pandemic. Noteworthy findings for the Chinese population included a striking increase in the proportion of Chinese people who felt “less safe” from 2018 to 2022, despite that Chinese people and other Ethnic Communities populations consistently experienced less crime than the national average during this period (NZCVS data).

Last modified: