Crosswords

Geographically Varying Effects of the Irish Famine: A Statistical Exploration

Description
Geographically Varying Effects of the Irish Famine: A Statistical Exploration
Categories
Published
of 14
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  Drew Webster Geographically Varying Effects of the Irish Famine: A Statistical Exploration I NTRODUCTION   The tragic mid-nineteenth century events surrounding  An Ghorta Mór—  otherwise known as the Irish Potato Famine—forever changed the people, objects, and landscape of Ireland. However, the Famine did not affect all citizens equally; the rural poor of Ireland’s south and west were the hardest hit (Miller 1985). This project uses historic census data to explore the geographic differences in Famine-period demographics to see what statistically significant  patterns emerge. I focus on data from two geographic poles: County Cork in the far southwest corner of the island of Ireland, and County Antrim, in the northeast corner of what is now  Northern Ireland (Figure 1). While the urban centers of Cork City, County Cork, and Belfast, County Antrim, both saw population increases over the 1840s, the rural areas of the county were affected quite differently by the Famine. Rural Cork saw heavy losses in population due to death and emigration, while in rural Antrim, the effects on the population were more minor (Smyth 2012a). This difference may be explained by the fact that the potato was a dietary staple in Cork, whose residents were mostly farmers, while Oatmeal was preferred in Antrim, whose residents were heavily involved in industry and manufacturing (Miller 1985; Smyth 2012b). Furthermore, the population in Antrim was generally wealthier and more highly educated than their Cork counterparts (Smyth 2012b). Here, I will take previously published census data from Bushell (2007) and reanalyze it to see if I can find statistically significant patterns in the demographic data. M ATERIALS AND M ETHODS   Due to a 1922 fire at The Public Records Office in Dublin, Irish census records from the nineteenth century are notoriously incomplete, but this has not stopped scholars from researching demographic changes before and after the Famine. Bushell (2007) analyzes demographic  patterns between the 1841 and 1851 censuses in County Cork and County Antrim focusing  primarily on death rate, but also considering ages and occupations. His data is publically available online as a comma-delimited text file. The surviving County Antrim data from 1851 include part or all of 14 parishes out of a total of 77 in the county, representing 8.7% of the total  population of the county. In County Cork, the records from four parishes survive, accounting for only 0.6% of the total population of the county. This means that an analysis of death rate (calculated as number of death records over number of occupant records) is impossible since the  proportion of the surviving records is not an accurate representation of actual demographic data at the time. Four types of records were collected: occupant records, death records, absence records, and uninhabited records. A summary of these records is presented in Table 1. The occupant records contain at maximum twelve types of information: County name, parish name, townland name, occupant name, age, sex, relationship to head of household, marital status, year of marriage, occupation, place of birth, and literacy. The absentee records contain the same information but apply to members of the household not currently residing there at the time of the census collection. Death records also record the same types of information, with the addition of cause of death, season of death, and year of death. Vacancy records merely note vacant households at the time of census collection. Importantly, death records only survive for County  Antrim, so an in-depth cross-country comparison of death records is not possible with the current data. However, other demographic differences between the two counties at the time of the Famine can still be explored. Furthermore, geographic differences in death within County Antrim can be analyzed at the parish (sub-county) level (Figure 2). A PPROACH   In order to understand what geographic and demographic differences existed in Ireland in the mid-nineteenth century,   I will test four hypotheses structured on three overlapping demographic themes: age, sex, and chronology. Age If the effects of the Famine vary geographically, does the age at death vary geographically as well? Furthermore, if the effects of the Famine vary geographically, we would expect to see a difference in age distributions between County Antrim and County Cork, especially if it affected the young or old more. These are Hypotheses 1 and 2: H 1 : The age of death varies geographically by parish in County Antrim. H 0 : The age of death is the same across all County Antrim parishes. H 2 : The age distribution of those alive for the 1851 census varies geographically between County Antrim and County H 0 : The age distribution is the same in both County Antrim and County Cork. Sex Does the Famine affect men and women differently in County Antrim? This leads to Hypothesis 3: H 3 : Death rates vary between men and women in County Antrim. H 0 : Death rates were the same for both men and women in County Antrim. Chronology What years and seasons did the Famine hit hardest in County Antrim? This is Hypothesis 4: H 4 : Death rates vary across years and seasons in County Antrim. H 0 : Death rates are the same for all years and seasons of death in County Antrim. R  ESULTS   Several statistical tests were performed to test these hypotheses. The results are below,  presented thematically.   Age H 1 : The age of death varies geographically by parish in County Antrim. A histogram presenting the age at death for villagers in County Antrim is presented in Figure 3. This shows that children under 10 years of age were the highest casualties of the Famine. When the bin numbers are increased to show detail (Figure 4), we see that over 15% of casualties were less than one year of age, demonstrating a high infant mortality rate. But does the age of death vary geographically?  An analysis of variance showed that the effect of parish on age at death was significant, F(13, 2733) = 4.643, p< 0.001.) Post hoc analyses using Tukey’s criterion for significance indicated that variance was most significant (p < .001)between Killead (M=38.3, SD=28.8) and Ahoghill (M=24.4 SD=26.1), and Larne(M=27.3, SD=28.5) and Killead (M=38.3, SD=28.8). Thus, the mean age at death does vary geographically among parishes in County Antrim, although this variance is more prevalent between some parishes than others. The results are  presented in Figure 5. H 2 : The age distribution of those alive for the 1851 census varies geographically between County Antrim and County Cork.   Histograms presenting the age distributions of citizens alive for the 1851 census from County Cork and County Antrim are presented in Figure 6. At the county level, these distributions appear similar, as do their means (Figure 7) but are there variations in the mean ages when we go to the parish level? When the mean ages of all recorded parishes from both County Cork and County Antrim were analyzed, an analysis of variance showed that the effect of parish on age was significant, F(17, 27879) = 4.047, p< 0.001.) Post hoc analyses using Tukey’s criterion for significance indicated that variance was most significant (p < .001)between Ahoghill, Antrim (M=22.7, SD=18.6), and several other villages, including Aghalee, Antrim (M=27.0, SD=19.5), Ballinderry, Antrim (M=26.2, SD=19.3), Killead, Antrim (M=26.7, SD=20.2), and Kilworth, Cork (M=26.2, SD=18.6), Thus, the mean age of villagers does vary geographically among  parishes in Counties Antrim and Cork, although this variance is more prevalent between some  parishes than others, and there is much variance between parishes in the same county. The results are presented in Figure 8. Sex H 3 : Death rates vary between men and women in County Antrim. On first glance, the Famine appears to have affected both men and women equally. Of all 2744 death records for which a biological sex is recorded, 51.7% are men and 48.3% are women. But does this differ geographically? To find out, a chi-square test of independence was  performed to examine the relation between sex of the deceased and parish. The relation between these results was not significant, X2 (13, N = 2747) = 7.63, p = 0.87. Thus number of dead by gender does not vary geographically by parish in County Antrim. The results are presented in Table 2. Perhaps men and women died at different ages? A Wilcoxon Rank Sum Test was  performed to analyze the relationship between age at death and sex. The relation between these variables was not significant, W=933396, Male = 25, Female = 25, p = 0.26. In fact, the age that County Antrim citizens died during the Famine does not vary significantly by sex. Maybe men and women died at different times of the year? A chi-square test of independence was performed to examine the relation between sex and season of death. The relation between these results was not significant, X2 (3, N = 2747) = 2.31, p = 0.51. The season that County Antrim citizens died during the Famine does not vary significantly by sex. All in all, it appears that men and women were equally affected by the Famine.  Chronology H 4 : Death rates vary across years and seasons in County Antrim. A histogram of deaths in County Antrim (Figure 9) shows that Famine deaths in County Antrim peaked in 1847. Deaths varied by season: 30% of deaths occurred in the winter, 30% in spring, 24% in summer, and 16% in autumn. Does the season of death vary by year? In County Antrim, the year of death (1829-1851) was subjected to a two-way analysis of variance having fourteen parishes and four seasons. Results showed that the year of death, F(13, 2730) = 2.12, p = 0.011, varied significantly by parish. Post hoc analyses using Tukey’s criterion for significance indicated that variance was most significant between the parishes of Carncastle (M=1845.8, SD=3.0) and Ahoghill (M=1847.0, SD=2.4), p<0.01, and Rasharkin (M=1846.9, SD=2.4) and Carncastle (M=1845.8,SD=3.0), p=0.0243. The year of death, F(3, 2740) = 7.99, p < 0.01, also varied significantly by the season of death, indicating that more people who died in the summer were more likely to do so during the early years of the Famine, and those who died in the winter or spring were more likely to do so during the later years of the Famine, with 1847  being the peak year of death for all seasons. This was most significant between summer deaths (M=1846.1, SD=2.6) and spring deaths (M=1846.6, SD=2.8), p=.002, and between summer deaths (M=1846.1, SD=2.6) and winter deaths(M=1846.7, SD=2.8), p<0.001. Furthermore, the interaction of both parish and season of death on year of death, F(38, 2689) = 1.44, p = .039, was also significant. The results are presented in Figure 10. I NTERPRETATION   The results give us many insights into the details and demographics of the Irish Famine. Hypothesis 1, that the age of death varies geographically in County Antrim, has been supported  by the data. Infant mortality is key here, since the deaths of infants make up a large portion of the overall dead. Hypothesis 2 has also been supported: age distributions do vary between parishes in Cork and Antrim, However, this doesn’t necessarily mean that the overall age distributions vary greatly at the county level—this variation was noted primarily at the parish level. Thus, the differences in age distribution may fluctuate broadly in villages within the same county, as opposed to there being significant differences between counties. The null hypothesis for Hypothesis 3 cannot be rejected. Thus, it appears that the Famine affected both men and women equally, regardless of location, age, or season of death. Finally, Hypothesis 4 has been supported: deaths from the Famine vary by year and season. In County Antrim, deaths peaked in 1847. More people died in the winter or spring than in other seasons; this may be attributed to the harsh weather and lack of food during those times causing diseases to take hold and spread.   C ONCLUSION     Statistical analysis has shown that the Famine did not affect all people or areas equally. Variations exist at the parish and county level, and death tallies vary by year and season. Overall, there is great potential for the statistical analysis of historic census data, but these results will only truly be significant if the dataset is complete.  F IGURES   !"#$%& () *+$,-. /01 +2 3%&40,5 6"-7 8,-%"9 0,5 *+%: ", ;&5 <=+$%>&) ?,"@&%A"-. *+44&#& B$C4",D
Search
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x