Table 1:

Distribution of covariates in the original, unweighted sample of households in Nunavut before the launch of Nutrition North Canada, during implementation with a 1-year lag and after full implementation

Covariate	% of households^*				p value^†
Covariate	Total sample (2007–2016) n = 3250	Before launch (2007–2010) n = 1360	Implementation with 1-year lag (2011–2013) n = 990	After full implementation (2014–2016) n = 900	p value^†
Household type^‡
Couple without children	11	12	12	9	0.08
Couple with children^§	48	47	46	51	0.1
Single person	24	25	25	24	0.9
Single female parent^§	12	12	12	12	0.9
Other	5	4	5	5	0.7
Presence of child < 18 yr	59	58	56	62	0.02
Respondent identified as Inuit	71	71	71	71	> 0.9
Household income,^¶ mean ± SD, ^$	49 015 ± 47 325	46 895 ± 40 795	51 320 ± 49 165	49 690 ± 53 940	0.06
Household income ≤ LIM	37	35	38	38	0.2
Imputed household income	33	33	31	37	0.04
Main income source
Wage, salary and self-employment	75	76	75	72	0.1
Social assistance	10	10	9	12	0.2
Other	15	14	16	16	0.3
Any social assistance^**	23	20	21	29	< 0.001
Any employment insurance^**	8	6	8	10	< 0.001
Any seniors’ income^**	11	11	12	8	0.001
Highest household education
Less than high school	20	18	22	20	0.07
High school completed	14	13	16	14	0.2
Postsecondary completed	56	56	54	58	0.2
Missing	10	13	8	8	< 0.001
Household owns dwelling	27	28	26	27	0.5
Respondent aged < 18 yr	14	15	15	12	0.08

Note: LIM = low income measure, SD = standard deviation.
↵* Unless stated otherwise. Column percentages may not add to 100% because of rounding.
↵† Obtained from χ² tests for proportions, or Wald tests for equality of means.
↵‡ All categories include other unrelated household members (e.g., couple without children alone or with others).
↵§ Includes children of all ages.
↵¶ Household income adjusted for household size by dividing by the square root of household size, and for inflation using the Consumer Price Index; expressed in 2007 constant dollars and rounded to the nearest 5 for presentation.
↵** Missing values grouped with the “no” category. All other covariates either had no missing values or had a “missing” category, or missing values were grouped with the “other” category.