MODULE 4 BIRTH TABULATIONS AND ANALYSIS GRAPHS Online

MODULE 4 BIRTH TABULATIONS AND ANALYSIS & GRAPHS Online course on data analysis and report writing for civil registration based vital statistics 12 July to 10 September 2021

RECAP: MANAGING DATA BIRTHS IN SMALL POPULATIONS v. When examining vital events in small populations, the numbers of births and deaths are likely to “jump around” a lot over time. v. To smooth out some of this noise and to see more clearly what is going on over time, we: v. Can calculate the average over a given period, or v. Can calculate rolling averages.

AVERAGING OVER A GIVEN PERIOD v. Calculate the average number of births over a 3 to 5 Year period v. Sum up all the births over 5 years and divide by 5. For example, if we had 20 birth occurring over 5 years (2008 -2012), we divide 20 by 5 to get 4 births per year. v. Do this for male and female babies separately. Example drawn from Tonga Vital Statistics report 2013 -2018

USING A ROLLING AVERAGE v. Rolling averages (also known as moving averages) can smooth out some of the noise so that any given point that is unusually high or low does not distort overall trends over time. v. We can average data points over 5 years and graph the midpoints of our data to see how things are changing. v. Helpful when policy makers request annual figures to understand changes in births or deaths over time

NIUE VITAL STATISTICS REPORT – ROLLING AVERAGE EXAMPLE

USING A ROLLING AVERAGE – GRAPH COMPARISON

IMPORTANCE OF BIRTH DATA v. Birth data has important statistical and administrative uses: It is key in the planning for infrastructure and delivery of social services. v Statistical uses: v. Measurement of population growth, fertility rates v. Providing denominators for early mortality calculations e. g. infant mortality rate (IMR), Neo natal mortality rate and under 5 mortality rate U 5 M. v Administrative uses: v. Planning infrastructure for: health, school, housing etc. v. Sub-national birth data facilitates following up children eligible for key health and social services e. g. education and immunization.

DEFINITION OF A LIVE BIRTH v. The WHO defines a live birth as the complete expulsion or extraction from its mother of a product of conception, irrespective of the duration of the pregnancy, which, after such separation, breathes or shows any other evidence of life e. g. beating of the heart, pulsation of the umbilical cord or definite movement of voluntary muscles - whether or not the umbilical cord has been cut or the placenta is attached. Each product of such a birth is considered live born. Much v. In more practical terms, a live birth is a birth where a newborn, nicer regardless of the length of his or her gestation, is born and demonstrates any sign of life.

STILL BIRTHS ARE NOT LIVE BIRTHS v. Should not be included in birth data for calculating vital statistical rates. v. A stillborn baby is a baby born after the 24 th week of pregnancy who does not show any signs of life. v. If the baby dies in the womb, it is known as an intra-uterine stillbirth. v. If the baby dies during labour, it is called an intra-partum stillbirth. v. If the baby dies before 24 weeks, it is known as a miscarriage.

UNDERSTANDING BIRTH DATA v. Suppose: v. Country A had 164, 000 births in 2011 v. Country B had 3, 000 births in 2011 v. Question: v. Which country has higher fertility?

ANSWER: WE DON’T KNOW v. The number of births alone does not tell us anything about the fertility of the average woman in a country. v. A large number of births may simply reflect a large population of mothers. Data in 2011 Country Population 2011 Births Total Fertility Rate Country A Country B Papua New Guinea Tonga 6, 188, 000 106, 000 164, 000 3. 5 3. 7 Data from the U. S. Census Bureau’s International Data Base

COMMON WAYS TO ANALYZE BIRTH DATA Some basic tabulations to describe our data: 1. Births by Age of the mother 2. Births by Place of birth 3. Births by Birth weight 4. Births by Gestation of pregnancies

BIRTHS BY AGE OF THE MOTHER v. Women of childbearing age are considered to be women aged between 15 -49 years. v. Births however do occur to mothers below 15 or older than 49 v It is important we capture these events, to better target interventions and service provision for these high-risk pregnancies v. Births per 1, 000 women aged 15 -19 is referred to as adolescent birth rate v Note that this does not include births to women under age 15 for comparability purposes. v The proportion of births to girls aged <15 and 15 -19 should be computed separately. Adolescent birth rate will be covered later with ASFRs

BIRTHS BY AGE OF THE MOTHER v. Age of mother at birth is important from a population growth perspective vhaving children at younger ages increases population growth even if young and old women have the same number of children v. The children that younger women are having now are able to have their own children sooner than the cohort of children born to those mothers when they are older vless time between generations

PROPORTION OF BIRTHS BY AGE OFExample THE MOTHER drawn from Nauru Vital Statistics report 2015 -2017

PLACE OF BIRTH v. Where are births occurring in your country? vhospitals, clinics, other health care centers, home births, or other places v. What percent were attended by a doctor or health care professional? v. How do births in your country compare to those occurring overseas v. Are there factors that are influencing mothers’ decisions on where to give birth from year to year? (see an example for Niue next slide) v. If the data supports it, how does this change over time?

PLACE OF BIRTH Example drawn from Niue Vital Statistics Report 2012 -2016

BIRTH WEIGHT v. To examine how many higher risk babies are being born, we look at the percentage of babies with low birth weights v. Determine the percent of births that are low-weight births, i. e. under 2500 grams (5. 5 lbs). Example drawn from Fiji Vital Statistics report 2012 - 2017

SEX RATIO AT BIRTH v. The sex ratio at birth is the ratio of male to female babies born. v. The normal sex ratio at birth for human babies is about 103106. vfactors influencing conception and intra-uterine mortality v. Sex ratio can vary naturally between different countries, cultures, and geographic locations

SEX RATIO AT BIRTH Slide figure from U. S. Census Bureau

SEX RATIO AT BIRTH v. Prenatal sex selections (in favor of sons) and decreasing fertility led to significant disparities in the ratio of males to females in these populations v. Technology allows parents to know baby’s sex from an early age, parents may selectively abort female fetuses. v. Aside from the ethical issues of sex selection, sociological problems can result v. Shortage of eligible women (compared to men) for marriage –effect on fertility v. Men behave differently than women – voting, economic purchasing, career tracks etc.

OTHER MEASURES OF FERTILITY - CBR v. Birth data are important v. As data in their own right for planning & Denominators for other rates v. But recall: births alone do not tell us about fertility of the average woman, and a large number of births may reflect a large population v. A common measure of fertility is the crude birth rate (CBR): vthe number of births per 1, 000 population over a given period of time.

CRUDE BIRTH RATE v. Why we use the CBR: veasy to understand vrequires the least amount of information vhelps users to understand fertility’s contribution to population growth

CRUDE BIRTH RATE IN SMALL POPULATIONS v. Generally, CBR = the number of births occurring in a year divided by the population at midyear, times 1, 000. v. If we aggregate over 5 years, divide the average number of births over this period by the population size at the midpoint of these 5 years. v. CBR of 5 year period = 1000 * (Average number of births over 5 years)/(Midpoint population)

CBR EXAMPLE v. Average number of births from 2003 -2007 is 325, we will need to divide this by our midpoint population. v. Our midpoint is July 1, 2005. Let’s assume the population was 15, 645 on July 1, 2005. We then perform the calculation: 1000 * (325/15, 645) to get a CBR of 20. 8 We can say there were 20. 8 births per 1, 000 population in 20032007.

QUESTION: v. Which country has higher fertility? v. Country X had a CBR of 29 in 2011 v. Country Y had a CBR of 25 in 2011

ANSWER: WE DON’T KNOW v. The CBR has some limitations as a measure of fertility. v. Misleading due to the composition of its denominator. vincludes children, men, and women outside of childbearing ages. v. The CBR is affected by the proportion of the population who cannot have children. v A very young or very old population or one that has many more males than females would affect the CBR. v Even when the frequency of having children among women of reproductive ages is the same in two countries, each country may have different crude birth rates.

POPULATION PYRAMID Population pyramids are a useful way of seeing the age and sex structure of our population. The CBR is larger in populations where women of reproductive age comprise a larger proportion of its people.

COMPARISON OF CBR BY PROPORTIONS OF THE POPULATION THAT ARE WOMEN OF CHILDBEARING AGE FOR 2 FICTITIOUS COUNTRIES Country A Country B Midpoint Population 10, 000 100, 000 Prop. that are women 15 -49 0. 30 0. 15 Total women aged 15 -49 3, 000 15, 000 Number of births* 3, 000 15, 000 CBR 300 150 *For calculation purposes, we assume all of these women had one child each in the year we are performing our calculation. I. E. Total fertility is the same between countries.

THE CBR IS LARGER IN POPULATIONS WHERE WOMEN OF REPRODUCTIVE AGE COMPRISE A LARGER PROPORTION OF Tonga RMI ITS PEOPLE. 15 10 5 Country X (RMI) had a CBR of 29 in 2011 but TFR=3. 4 0 5 10 15 Country Y (Tonga) had a CBR of 25 in 2011 but TFR=3. 7 Female 0 Male 1 0 8 0 9 8 5 + 0 7 0 + 7 5 8 6 0 8 9 6 5 7 4 5 0 7 9 5 5 6 4 0 6 9 4 5 5 4 3 0 5 9 3 5 4 4 2 0 4 9 2 5 3 4 1 0 3 9 1 5 2 4 0 2 9 5 1 4 0 1 9 9 4 4 Population data from the U. S. Census Bureau’s International Data Base

WHY DO WE CARE ABOUT THE CBR? v. Why we use the CBR: veasy to understand vrequires the least amount of information vhelps use understand fertility’s contribution to population growth v. We can standardize the CBR to compare between 2 populations (different countries or same country over time). v. More about this in the later sessions

A NOTE ABOUT THE MIDPOINT POPULATION v. Due to small population sizes and the relatively small number of occurrences, we will aggregate over 3 -5 year periods. v. To accurately calculate rates, the midpoint population should be used in the denominator. v For example, if you have births from 2008 -2012, the midpoint would be July 1, 2010. v The population on July 1, 2010 would be used in the denominator as this is exactly halfway between 2008 -2012. v. If you have midpoint pop from your NSO by sex and 5 year age group – use this. If not, you can use SPC projections if your midpoint falls between 20002020

ASSIGNMENT v. Please tabulate your live birth date by: 1. 2. 3. 4. Year of birth and sex Period of birth (combing 3 or 5 Years of data) Place of birth Age of mother Please compute the: 1. Crude birth rate 2. Sex Ratio at birth Depending of the number of years of data you have, you can do this by three or 5 year periods

GRAPHING YOUR DATA

WHY USE CHARTS? • Comparison: how much? Which item is bigger or smaller? • Changes over time: how does a variable evolve? • Frequency distribution: how are items distributed? What are the differences? • Correlation • Relative share of a whole

CHECKLIST FOR A GOOD CHART A good graph: • grab’s attention • Presents information clearly, simply, accurately • Does not mislead • Facilitates comparisons and highlights trends • Illustrates messages, themes or stories in text

BAR CHARTS

STACKED COLUMN GRAPHS

LINE GRAPHS

PIE CHARTS

SCATTER PLOT

ADVANTAGES AND DISADVANTAGES Chart type Advantages Disadvantages Bar chart (vertical) Simple and clear Works for categories and time series Not good for long time series Small space for long names Bar chart (horizontal) Good for large number of categories Works for long names Not appropriate for time series Line chart Simple and clear Best for time series More than three lines gets confusing Pie chart Shows distribution of one variable Not good for making comparisons Too many ‘slices’ gets confusing Scatter plots Shows relationships between variables Can be difficult to interpret

COMPONENTS OF A GOOD GRAPH

WHICH TYPE SHOULD I USE?

SORT YOUR DATA

RE-SETTING AXIS SCALES – BE CAREFUL

KEEP IT SIMPLE

KEEP IT SIMPLE

KEEP IT SIMPLE

BE CAREFUL ON LOAD OF INFORMATION

ANY QUESTIONS?