# A tibble: 6 × 9
Year Maternal Race or Ethnici…¹ Infant Mortality Rat…² Neonatal Mortality R…³
<dbl> <chr> <dbl> <dbl>
1 2016 Puerto Rican 3.4 2.4
2 2016 Asian and Pacific Islander 2.9 2
3 2016 Other/Two or More NA NA
4 2016 Non-Hispanic Black 8 4.9
5 2016 Non-Hispanic White 2.6 1.6
6 2016 Other Hispanic 3.8 2.4
# ℹ abbreviated names: ¹`Maternal Race or Ethnicity`, ²`Infant Mortality Rate`,
# ³`Neonatal Mortality Rate`
# ℹ 5 more variables: `Postneonatal Mortality Rate` <dbl>,
# `Infant Deaths` <dbl>, `Neonatal Infant Deaths` <dbl>,
# `Postneonatal Infant Deaths` <dbl>, `Number of Live Births` <dbl>
Introduction
Infant Mortality Rate is defined as the number of deaths per 1,000 live births that occur prior to the infant reaching the age of 1 years old.
As time has passed the mean infant mortality rate in the U.S. has steadily declined among the general population, however observations have been made of infant mortality rates being disproportionately distributed among various different ethnic groups.
In order to determine whether these disparities are a result of racial inequality or are due to random chance I will be utilizing data gathered from a study performed by the CDC. The data for this study was collected through the use of randomized sampling & was performed by a very reputable organization, thus the results of this study were unlikely to be biased.
The statistic techniques that I plan to use to determine whether or not Race/Ethnicity affects the infant mortality rate of a population are Chi-Square tests, linear regressions, & ANOVA.
The overarching question I would like to answer is whether or not infant mortality rates in America are independent of maternal Race/Ethnicity.
Data
After loading my Dataset I used this chunk to edit my subtitles, making them lowercase & replacing spaces with underlines.
# A tibble: 6 × 9
year maternal_race_or_ethnicity infant_mortality_rate neonatal_mortality_rate
<dbl> <chr> <dbl> <dbl>
1 2016 Puerto Rican 3.4 2.4
2 2016 Asian and Pacific Islander 2.9 2
3 2016 Other/Two or More NA NA
4 2016 Non-Hispanic Black 8 4.9
5 2016 Non-Hispanic White 2.6 1.6
6 2016 Other Hispanic 3.8 2.4
# ℹ 5 more variables: postneonatal_mortality_rate <dbl>, infant_deaths <dbl>,
# neonatal_infant_deaths <dbl>, postneonatal_infant_deaths <dbl>,
# number_of_live_births <dbl>
I then used this chunk to filter out any N/A’s present in my data to ensure that it can be read smoothly.
IM <- infant |>filter(!is.na(number_of_live_births)) |>filter(!is.na(infant_mortality_rate)) |>filter(!is.na(neonatal_mortality_rate)) |>filter(!is.na(postneonatal_mortality_rate)) |>filter(!is.na(infant_deaths)) |>filter(!is.na(postneonatal_infant_deaths)) |>filter(!is.na(year)) IM
# A tibble: 48 × 9
year maternal_race_or_ethnicity infant_mortality_rate neonatal_mortality_r…¹
<dbl> <chr> <dbl> <dbl>
1 2016 Asian and Pacific Islander 2.9 2
2 2016 Non-Hispanic Black 8 4.9
3 2016 Non-Hispanic White 2.6 1.6
4 2016 Other Hispanic 3.8 2.4
5 2015 Puerto Rican 6.1 4.5
6 2015 Non-Hispanic White 2.7 1.8
7 2015 Non-Hispanic Black 8 4.8
8 2015 Asian and Pacific Islander 2.6 1.6
9 2015 Other Hispanic 4.3 2.9
10 2014 Asian and Pacific Islander 2.6 1.8
# ℹ 38 more rows
# ℹ abbreviated name: ¹neonatal_mortality_rate
# ℹ 5 more variables: postneonatal_mortality_rate <dbl>, infant_deaths <dbl>,
# neonatal_infant_deaths <dbl>, postneonatal_infant_deaths <dbl>,
# number_of_live_births <dbl>
I will use this histogram to visualize the relationship between the # of infant deaths observed & their maternal ethnicity which can help illustrate whether certain races are more likely to die as infants.
ggplot(IM, aes(x= maternal_race_or_ethnicity, y= infant_deaths))+geom_histogram(stat ="identity") +theme(axis.text.x =element_text(size =4)) +ggtitle("Infant Deaths by Maternal Race or Ethnicity") +xlab("Maternal Race or Ethnicity") +ylab("Infant Deaths")
Warning in geom_histogram(stat = "identity"): Ignoring unknown parameters:
`binwidth`, `bins`, and `pad`
Using this chunk I plan to create a dot plot that helps to visualize the relative infant mortality rates of different ethinicities in a way that is easily digestible.
ggplot(IM, aes(x= maternal_race_or_ethnicity, y= infant_mortality_rate)) +geom_point(stat ="identity") +theme(axis.text.x =element_text(size =4)) +ggtitle("Infant Mortality by Maternal Race or Ethnicity") +xlab("Maternal Race or Ethnicity") +ylab("Infant Mortality")
With this chunk I plan create a scatterplot & to plot a line of best fit for the number of infant deaths to number of live births, color coded by the maternal race/ethnicity to see whether race results in any significant outliers to the best fit line.
ggplot(IM, aes(x= number_of_live_births, y= infant_deaths, colour = maternal_race_or_ethnicity)) +geom_point(stat ="identity") +theme(axis.text.x =element_text(size =4)) +geom_smooth(method ="lm", se =TRUE, colour="red") +ggtitle("Infant Live Births by Infant Deaths") +xlab("# of Live Births") +ylab("# of Infant Deaths")
`geom_smooth()` using formula = 'y ~ x'
summary(IM)
year maternal_race_or_ethnicity infant_mortality_rate
Min. :2007 Length:48 Min. : 2.600
1st Qu.:2009 Class :character 1st Qu.: 3.100
Median :2011 Mode :character Median : 4.300
Mean :2011 Mean : 5.062
3rd Qu.:2014 3rd Qu.: 6.650
Max. :2016 Max. :10.200
neonatal_mortality_rate postneonatal_mortality_rate infant_deaths
Min. :1.60 Min. :0.600 Min. : 46.0
1st Qu.:2.10 1st Qu.:0.975 1st Qu.: 62.0
Median :2.70 Median :1.600 Median :111.5
Mean :3.31 Mean :1.754 Mean :119.4
3rd Qu.:4.55 3rd Qu.:2.425 3rd Qu.:144.0
Max. :6.50 Max. :3.800 Max. :287.0
neonatal_infant_deaths postneonatal_infant_deaths number_of_live_births
Min. : 33.00 Min. : 12.00 Min. : 7561
1st Qu.: 43.00 1st Qu.: 20.75 1st Qu.:19372
Median : 75.00 Median : 36.50 Median :26230
Mean : 77.62 Mean : 41.73 Mean :25109
3rd Qu.: 97.00 3rd Qu.: 48.50 3rd Qu.:30104
Max. :182.00 Max. :110.00 Max. :40633
summary(IM$infant_mortality_rate)
Min. 1st Qu. Median Mean 3rd Qu. Max.
2.600 3.100 4.300 5.062 6.650 10.200
table (IM$maternal_race_or_ethnicity)
Asian and Pacific Islander Black Non-Hispanic
10 8
Non-Hispanic Black Non-Hispanic White
2 2
Other Hispanic Puerto Rican
10 8
White Non-Hispanic
8
Pearson's Chi-squared test with Yates' continuity correction
data: IMR
X-squared = 45.66, df = 1, p-value = 1.407e-11
With a p-value of 1.407e-11 I concluded that these values were very significant.
IM |>group_by(maternal_race_or_ethnicity) |>summarise(mean(infant_mortality_rate))
# A tibble: 7 × 2
maternal_race_or_ethnicity `mean(infant_mortality_rate)`
<chr> <dbl>
1 Asian and Pacific Islander 2.99
2 Black Non-Hispanic 8.81
3 Non-Hispanic Black 8
4 Non-Hispanic White 2.65
5 Other Hispanic 4.38
6 Puerto Rican 6.59
7 White Non-Hispanic 3.1
mod <-aov(infant_mortality_rate ~ maternal_race_or_ethnicity, data = IM)summary(mod)
Df Sum Sq Mean Sq F value Pr(>F)
maternal_race_or_ethnicity 6 238.42 39.74 155.6 <2e-16 ***
Residuals 41 10.47 0.26
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Based off of my P value of 2e-16 I would conclude that there is a statistically significant relationship between race & infant mortality rate.
Conclusions
Here I will add a general conclusion, restate any important results from my findings, & state my thoughts on the implications, as well as the validity of my findings
Work Cited
-Mortality rate, infants(per 1,000 live births). (2022). World Bank Open Data. https://data.worldbank.org/indicator/SP.DYN.IMRT.IN
-Infant mortality rate. (2024). We are the Nation’s first line of defense - CIA. https://www.cia.gov/the-world-factbook/field/infant-mortality-rate/country-comparison/
-Infant mortality rates by race/ethnicity: United States, 2018-2020 average. (n.d.). March of Dimes | PeriStats. https://www.marchofdimes.org/peristats/data?reg=99&top=6&stop=92&lev=1&slev=1&obj=1
-Infant mortality. (2024, May 20). Maternal Infant Health. https://www.cdc.gov/maternal-infant-health/infant-mortality/index.html
-Infant mortality in the United States, 2021: Data from the period linked birth/Infant death file.(September).PubMed.https://pubmed.ncbi.nlm.nih.gov/37748084/
