Spatial variation in the use of reproductive health services over time: a decomposition analysis

Table 1 Summary statistics for dependent and independent variables – pooled data for 1998 and 2014

Variables	Obs.	Mean	SD	Variables	Obs.	Mean	SD
4+ Antenatal Visits				Ethnicity
No	6386	0.214	0.410	Akan (Ref)	6386	0.399	0.490
Yes	6386	0.786	0.410	Ga/Dangme	6386	0.052	0.222
Woman’s Age	6386	30.493	7.229	Ewe and Guans	6386	0.136	0.343
Woman’s Age Sq.	6386	982.076	460.205	North ethnicities	6386	0.393	0.488
Birth order				Others	6386	0.020	0.140
One child (Ref)	6386	0.214	0.410	Number of elderly	6386	1.340	0.665
Two children	6386	0.193	0.395	Household wealth
Three children	6386	0.161	0.368	Poorest (Ref)	6386	0.315	0.465
Four and above	6386	0.431	0.495	Poorer	6386	0.214	0.410
Woman’s education				Middle	6386	0.183	0.387
No educ (Ref)	6386	0.374	0.484	Richer	6386	0.157	0.364
Primary	6386	0.193	0.395	Richest	6386	0.130	0.336
Secondary	6386	0.404	0.491	Ecological zones
Tertiary	6386	0.029	0.168	Southern belt (Ref)	6386	0.283	0.451
Partner education				Greater Accra	6386	0.085	0.279
No educ (Ref)	6386	0.296	0.457	Middle belt	6386	0.308	0.462
Primary	6386	0.100	0.300	Northern belt	6386	0.323	0.468
Secondary	6386	0.459	0.498	Rural dummy	6386	0.651	0.477
Tertiary	6386	0.074	0.261	NSCPHGW	6386	0.663	0.340
Missing Partners	6386	0.071	0.256	NSCPHFT	6386	0.100	0.209
Muslim	6386	0.298	0.458	Sample Dummy
Year Dummy				Skilled Attendance at Birth	6665
Year 1998	2242
Year 2014	4144

Source: Authors’ calculations. Calculations take account of sample weights. Note that the models on skilled attendants at birth is based on slightly different sample per the sample dummy. NSCPHGW and NSCPHFT are the non-self-cluster proportion of households with good water and non-self-cluster proportion of households with flush toilets respectively. Note, partner’s education includes a 5th category (missing partners). This was added to cater for women who do not have partners and would otherwise have been dropped from the regressions. The age variable has also been categorized with number of observations and percentages as follows: 15–19 (2645: 18.6%), 20–24 (2958: 17.3%), 25–29 (2421: 17%), 30–34 (2004: 14.1%), 35–39 (1887: 13.3%), 40–44 (1516: 10.7%) and 45–49 (1308: 9.2%)

ISSN: 1471-2393