Read in the Dataset

I have spent HW3 and HW4 cleaning up the 2017 Australian Marriage Law dataset and better understanding it. I returned to HW4 and attempted to save a cleaner version of the dataset as a .csv file using the readr function write_csv. I will try to read in the data below:

ausmar <- read_csv("/Users/Megan Georges/Documents/DACSS601_R/Data/ausmar.csv")

## Rows: 150 Columns: 6

## -- Column specification --------------------------------------------------------
## Delimiter: ","
## chr (2): Divisions, Cities
## dbl (4): Response_Clear_Yes, Response_Clear_No, Response_Not_Clear, Non_Resp...

## 
## i Use `spec()` to retrieve the full column specification for this data.
## i Specify the column types or set `show_col_types = FALSE` to quiet this message.

head(ausmar)

## # A tibble: 6 x 6
##   Divisions                 Cities    Response_Clear_~ Response_Clear_~ Response_Not_Cl~
##   <chr>                     <chr>                <dbl>            <dbl>            <dbl>
## 1 New South Wales Divisions Banks                37736            46343              247
## 2 New South Wales Divisions Barton               37153            47984              226
## 3 New South Wales Divisions Bennelong            42943            43215              244
## 4 New South Wales Divisions Berowra              48471            40369              212
## 5 New South Wales Divisions Blaxland             20406            57926              220
## 6 New South Wales Divisions Bradfield            53681            34927              202
## # ... with 1 more variable: Non_Response <dbl>

Fix Division Names… Again

I don’t like that each division name ends with “Divisions” instead of “Division”, so I used the mutate() function to change the names… Then I decided that having “Division” in each value of the Division column was repetitive, so I removed it altogether.

ausmar <- ausmar %>%
  mutate(Divisions = case_when(
    Divisions == "New South Wales Divisions" ~ "New South Wales",
    Divisions == "Victoria Divisions" ~ "Victoria",
    Divisions == "Queensland Divisions" ~ "Queensland",
    Divisions == "South Australia Divisions" ~ "South Australia",
    Divisions == "Western Australia Divisions" ~ "Western Australia",
    Divisions == "Tasmania Divisions" ~ "Tasmania",
    Divisions == "Northern Territory Divisions" ~ "Northern Territory",
    Divisions == "Australian Capital Territory Divisions" ~ "Australian Capital Territory"
  ))

Practice Generating and Designing a Table

I have really only been displaying data in tibbles. I wanted to play around with creating a more appealing and clear table for the dataset when the file is knitted. I decided to use the function kable() from the knitr package and download kableExtra because it is simple and works with a rectangular dataset like the cleaned-up Australian Marraige Law dataset.

kable(ausmar, col.names = c("Division", "City", "Response Yes (Clear)", "Response No (Clear)", "Response Not Clear", "No Response"), 
      align = c('c', 'c', 'c', 'c', 'c', 'c')) %>%
  add_header_above(c("Location"=2, "Type of Response"=4))%>%
    kable_styling(fixed_thead = TRUE)%>%
  scroll_box(width = "100%", height = "500px")

Location		Type of Response
Division	City	Response Yes (Clear)	Response No (Clear)	Response Not Clear	No Response
New South Wales	Banks	37736	46343	247	20928
New South Wales	Barton	37153	47984	226	24008
New South Wales	Bennelong	42943	43215	244	19973
New South Wales	Berowra	48471	40369	212	16038
New South Wales	Blaxland	20406	57926	220	25883
New South Wales	Bradfield	53681	34927	202	17261
New South Wales	Calare	54091	35779	285	25342
New South Wales	Chifley	32871	46702	263	28180
New South Wales	Cook	47505	38804	229	18713
New South Wales	Cowper	57493	38317	315	25197
New South Wales	Cunningham	60906	31840	268	20607
New South Wales	Dobell	59475	30987	255	24275
New South Wales	Eden-Monaro	57223	30926	249	22139
New South Wales	Farrer	48432	39297	277	25669
New South Wales	Fowler	27847	48782	228	29251
New South Wales	Gilmore	59322	36386	303	23109
New South Wales	Grayndler	73208	18429	136	16074
New South Wales	Greenway	38016	43980	217	25253
New South Wales	Hughes	51337	36558	185	17038
New South Wales	Hume	51284	36271	213	23457
New South Wales	Hunter	59137	32723	251	25253
New South Wales	Kingsford Smith	56297	31510	225	22399
New South Wales	Lindsay	49071	38295	234	26955
New South Wales	Lyne	51416	41539	316	21426
New South Wales	Macarthur	43323	39907	228	27271
New South Wales	Mackellar	62350	29330	208	17500
New South Wales	Macquarie	56180	31778	235	18490
New South Wales	McMahon	29146	53967	242	23721
New South Wales	Mitchell	42112	43652	176	19436
New South Wales	Newcastle	71158	23999	232	19970
New South Wales	New England	44608	40324	256	25581
New South Wales	North Sydney	64813	25473	193	17538
New South Wales	Page	55943	37727	291	25645
New South Wales	Parkes	41408	37108	241	29777
New South Wales	Parramatta	29299	47038	197	25757
New South Wales	Paterson	60915	32059	279	24264
New South Wales	Reid	43567	39061	203	23786
New South Wales	Richmond	62591	29625	274	22719
New South Wales	Riverina	47333	39308	265	25632
New South Wales	Robertson	58689	30614	231	20431
New South Wales	Shortland	62455	29836	255	19675
New South Wales	Sydney	76144	14860	146	22093
New South Wales	Warringah	64999	21660	172	16630
New South Wales	Watson	24915	57160	205	24634
New South Wales	Wentworth	69279	16410	162	18121
New South Wales	Werriwa	30252	53174	269	29282
New South Wales	Whitlam	57562	34879	276	23064
Victoria	Aston	48455	29730	234	17664
Victoria	Ballarat	65613	27405	333	20923
Victoria	Batman	66383	26906	287	17901
Victoria	Bendigo	63412	28852	333	19360
Victoria	Bruce	34644	39203	257	21261
Victoria	Calwell	37839	49823	331	23588
Victoria	Casey	59959	28144	316	16807
Victoria	Chisholm	49448	30844	271	17188
Victoria	Corangamite	69723	27708	326	17123
Victoria	Corio	62658	29865	359	18255
Victoria	Deakin	55464	28973	276	15389
Victoria	Dunkley	62840	24471	285	19322
Victoria	Flinders	68291	29275	336	21407
Victoria	Gellibrand	62045	29065	278	19771
Victoria	Gippsland	51196	33910	338	20370
Victoria	Goldstein	69726	21663	238	14857
Victoria	Gorton	49834	43587	347	27479
Victoria	Higgins	70059	19375	180	16615
Victoria	Holt	47147	45875	289	28260
Victoria	Hotham	47986	32524	303	19732
Victoria	Indi	54563	31925	324	18934
Victoria	Isaacs	56645	30063	275	20692
Victoria	Jagajaga	65098	23453	255	15363
Victoria	Kooyong	63592	22729	231	14147
Victoria	Lalor	57062	43429	345	30127
Victoria	La Trobe	61807	29826	268	19002
Victoria	Mallee	42495	35795	359	21207
Victoria	Maribyrnong	53208	35658	360	23762
Victoria	McEwen	73705	39007	377	26966
Victoria	McMillan	61479	36500	372	22403
Victoria	Melbourne	81287	15839	182	20154
Victoria	Melbourne Ports	70589	15523	198	18745
Victoria	Menzies	47137	35626	258	15745
Victoria	Murray	48205	35452	357	21560
Victoria	Scullin	48245	42147	357	22817
Victoria	Wannon	49340	31529	343	18569
Victoria	Wills	68450	29399	250	20169
Queensland	Blair	47194	31433	256	23975
Queensland	Bonner	52139	31891	209	17981
Queensland	Bowman	53529	32627	220	19691
Queensland	Brisbane	72812	18762	159	20656
Queensland	Capricornia	39917	33917	230	24896
Queensland	Dawson	42539	34599	205	26943
Queensland	Dickson	54206	28988	190	18527
Queensland	Fadden	52154	32218	215	26234
Queensland	Fairfax	58510	32451	277	21335
Queensland	Fisher	52023	30783	258	19625
Queensland	Flynn	39020	36783	262	24529
Queensland	Forde	46937	30585	231	24423
Queensland	Griffith	69171	21132	184	20133
Queensland	Groom	40536	41915	262	20717
Queensland	Herbert	48110	28441	207	29665
Queensland	Hinkler	40649	39548	308	22425
Queensland	Kennedy	33160	37784	263	29794
Queensland	Leichhardt	47750	27606	239	35841
Queensland	Lilley	59991	28671	218	20249
Queensland	Longman	51268	33576	258	24250
Queensland	Maranoa	35475	45308	352	22553
Queensland	McPherson	54034	28486	229	23214
Queensland	Moncrieff	50566	28717	214	25232
Queensland	Moreton	47418	30413	220	20020
Queensland	Oxley	44655	29365	224	23348
Queensland	Petrie	53144	33067	200	23323
Queensland	Rankin	41570	34621	237	26119
Queensland	Ryan	64967	24451	162	16223
Queensland	Wide Bay	46507	37065	319	21645
Queensland	Wright	47109	35812	280	22144
South Australia	Adelaide	62769	26771	217	20477
South Australia	Barker	42498	38827	243	24297
South Australia	Boothby	62139	28556	234	16919
South Australia	Grey	40811	35750	260	25327
South Australia	Hindmarsh	57947	33613	273	20548
South Australia	Kingston	58863	27567	252	20764
South Australia	Makin	51547	33743	242	21991
South Australia	Mayo	57361	31247	261	17239
South Australia	Port Adelaide	53649	33869	276	27253
South Australia	Sturt	52308	32655	252	19410
South Australia	Wakefield	52636	33649	268	27802
Western Australia	Brand	51953	25481	194	24466
Western Australia	Burt	44058	33275	169	24197
Western Australia	Canning	48486	32019	214	22157
Western Australia	Cowan	44388	31075	184	21330
Western Australia	Curtin	59638	22943	178	15706
Western Australia	Durack	39304	27128	194	31428
Western Australia	Forrest	51612	29285	225	21752
Western Australia	Fremantle	57541	24559	236	19878
Western Australia	Hasluck	47880	28836	230	19570
Western Australia	Moore	56690	26690	195	16916
Western Australia	O’Connor	43554	33987	234	24925
Western Australia	Pearce	54305	30699	209	26401
Western Australia	Perth	57510	22967	177	19479
Western Australia	Stirling	47225	30060	190	21345
Western Australia	Swan	49093	26830	185	21857
Western Australia	Tangney	48338	30090	174	14926
Tasmania	Bass	36249	22510	145	15487
Tasmania	Braddon	30054	25573	154	17632
Tasmania	Denison	45005	15992	167	13092
Tasmania	Franklin	44746	20322	163	13605
Tasmania	Lyons	35894	25258	176	17204
Northern Territory	Lingiari(c)	19026	15898	106	34854
Northern Territory	Solomon	29660	15792	123	22642
Australian Capital Territory	Canberra(d)	89590	31361	281	24399
Australian Capital Territory	Fenner(e)	85869	30159	253	26196

Proportion of Voters that Did Not Respond

I think that a very useful piece of information for researchers and organizers to pay mind to is the proportion of voters that did not respond to the marriage law survey. I’d be curious to learn more about what produced high or low rates of response in different areas.

ausmartotals <- ausmar %>%
  mutate(Total = (Response_Clear_Yes + Response_Clear_No + Response_Not_Clear + Non_Response))

ausmarnon <- ausmartotals %>%
  select(Divisions, Cities, Non_Response, Total) %>%
  mutate(Prop_Non_Response = (Non_Response/Total *100))

ausmarnon %>%
  select(Divisions, Cities, Prop_Non_Response) %>%
  arrange(desc(Prop_Non_Response)) %>%
  kable(col.names = c("Division", "City", "Non Response Proportion"), 
        align = c('c', 'c', 'c'), 
        caption = "Proportion of Voters that Did Not Respond in Each City") %>%
  kable_styling(fixed_thead = TRUE)%>%
  scroll_box(width = "100%", height = "500px")

Proportion of Voters that Did Not Respond in Each City
Division	City	Non Response Proportion
Northern Territory	Lingiari(c)	49.87408
Northern Territory	Solomon	33.19114
Queensland	Leichhardt	32.16286
Western Australia	Durack	32.05173
Queensland	Kennedy	29.49872
Queensland	Herbert	27.87461
New South Wales	Fowler	27.56720
New South Wales	Parkes	27.43564
New South Wales	Chifley	26.08873
New South Wales	Werriwa	25.91855
Queensland	Dawson	25.83568
Queensland	Rankin	25.47027
New South Wales	Parramatta	25.18012
Queensland	Capricornia	25.15764
South Australia	Grey	24.79442
New South Wales	Blaxland	24.78384
New South Wales	Macarthur	24.62860
Queensland	Flynn	24.38416
South Australia	Wakefield	24.31201
Western Australia	O’Connor	24.26972
Queensland	Moncrieff	24.09266
Tasmania	Braddon	24.01754
Western Australia	Brand	23.96419
Queensland	Oxley	23.92409
Queensland	Forde	23.90287
Western Australia	Burt	23.79276
South Australia	Port Adelaide	23.68858
Queensland	Fadden	23.67241
Western Australia	Pearce	23.65384
New South Wales	Lindsay	23.53018
New South Wales	Greenway	23.49859
Queensland	Blair	23.30883
Victoria	Holt	23.24568
New South Wales	New England	23.09401
New South Wales	Watson	23.04095
Victoria	Lalor	23.00421
South Australia	Barker	22.95093
New South Wales	Riverina	22.77631
Victoria	Gorton	22.66365
New South Wales	Farrer	22.58104
Western Australia	Swan	22.31103
New South Wales	Reid	22.30976
Victoria	Bruce	22.29434
Queensland	Longman	22.17609
New South Wales	McMahon	22.15342
Western Australia	Cowan	21.99491
New South Wales	Barton	21.95097
New South Wales	Calare	21.94170
Queensland	McPherson	21.90765
Tasmania	Lyons	21.90699
Queensland	Hinkler	21.78665
Queensland	Maranoa	21.75083
Western Australia	Stirling	21.59988
Western Australia	Canning	21.53758
New South Wales	Hunter	21.51682
New South Wales	Page	21.44123
Queensland	Petrie	21.25412
Victoria	Mallee	21.23758
Western Australia	Forrest	21.14431
Victoria	Calwell	21.13980
New South Wales	Dobell	21.11016
New South Wales	Hume	21.08968
Victoria	Maribyrnong	21.03055
Queensland	Wright	21.02046
Tasmania	Bass	20.81838
New South Wales	Cowper	20.76870
New South Wales	Paterson	20.64723
Queensland	Wide Bay	20.50959
South Australia	Makin	20.45237
Victoria	Murray	20.42169
Queensland	Moreton	20.41378
New South Wales	Kingsford Smith	20.28325
Western Australia	Hasluck	20.27643
Victoria	Scullin	20.09140
Queensland	Groom	20.02997
New South Wales	Eden-Monaro	20.02859
New South Wales	Whitlam	19.92037
New South Wales	Banks	19.88333
New South Wales	Richmond	19.71981
Victoria	Hotham	19.62504
New South Wales	Sydney	19.50937
Western Australia	Perth	19.45313
Western Australia	Fremantle	19.44743
New South Wales	Gilmore	19.39976
South Australia	Kingston	19.32506
Victoria	McEwen	19.25386
Victoria	Gippsland	19.25076
Victoria	Isaacs	19.21709
Queensland	Fisher	19.11110
Queensland	Fairfax	18.95215
New South Wales	Bennelong	18.77603
New South Wales	Lyne	18.68052
Victoria	Wannon	18.60976
New South Wales	Robertson	18.57955
South Australia	Adelaide	18.57594
Queensland	Bowman	18.56468
Queensland	Lilley	18.55510
Victoria	McMillan	18.55259
South Australia	Sturt	18.55197
New South Wales	Mitchell	18.44443
Australian Capital Territory	Fenner(e)	18.38613
Victoria	Aston	18.38411
Queensland	Brisbane	18.37902
Victoria	Ballarat	18.30950
South Australia	Hindmarsh	18.28423
Queensland	Griffith	18.20014
Queensland	Dickson	18.17959
New South Wales	Cunningham	18.13661
Victoria	Dunkley	18.07179
Victoria	Flinders	17.94249
Victoria	Indi	17.90517
Victoria	Melbourne Ports	17.84303
Victoria	Gellibrand	17.78623
New South Wales	Cook	17.77940
Tasmania	Denison	17.63090
Queensland	Bonner	17.59049
Victoria	Chisholm	17.58345
New South Wales	Shortland	17.53237
New South Wales	Wentworth	17.42873
New South Wales	Macquarie	17.33172
New South Wales	Newcastle	17.31118
Victoria	Bendigo	17.29235
Tasmania	Franklin	17.25734
Victoria	Melbourne	17.15789
Victoria	La Trobe	17.13389
Victoria	Wills	17.05364
Western Australia	Moore	16.83335
Australian Capital Territory	Canberra(d)	16.75399
Victoria	Corio	16.42567
New South Wales	Bradfield	16.27306
South Australia	Mayo	16.24665
New South Wales	North Sydney	16.23633
New South Wales	Hughes	16.20845
New South Wales	Warringah	16.07369
Victoria	Batman	16.05802
New South Wales	Mackellar	15.99810
Victoria	Casey	15.97229
Western Australia	Tangney	15.95886
Western Australia	Curtin	15.95085
Victoria	Menzies	15.94172
South Australia	Boothby	15.68782
Victoria	Higgins	15.64074
Victoria	Deakin	15.37332
Queensland	Ryan	15.33321
New South Wales	Berowra	15.26120
Victoria	Corangamite	14.90512
New South Wales	Grayndler	14.90445
Victoria	Jagajaga	14.74815
Victoria	Kooyong	14.04880
Victoria	Goldstein	13.95233

A Basic Bar Graph of Total Votes

A very basic bar graph can be useful to display the total number of votes from the survey, including clear Yes and No responses, as well as Unclear responses and Non-Responses. The graph clearly displays that more respondents support a change in law than those who do not support a change in law.

ausmarvotes <- ausmar %>%
  select(Response_Clear_Yes, Response_Clear_No, Response_Not_Clear, Non_Response)%>%
  summarise(Yes=sum(Response_Clear_Yes), No=sum(Response_Clear_No), NotClear=sum(Response_Not_Clear), NonResponse=sum(Non_Response))

rownames(ausmarvotes) <- c("Total")

## Warning: Setting row names on a tibble is deprecated.

ausmarvotes <- pivot_longer(ausmarvotes, Yes:NonResponse, names_to = "Response", values_to = "Total") %>%
  arrange(desc(Total)) 
kable(ausmarvotes)

Response	Total
Yes	7817247
No	4873987
NonResponse	3278260
NotClear	36686

ggplot(data = ausmarvotes) +
  geom_bar(mapping = aes(x = Response, y = Total), stat = "identity") +
  labs(title = "Should Australian Law Change to Allow Same-Sex Marraige?", 
       subtitle = "2017 postal survey of Australians", x = "Type of Response", y = "Total # of Responses")

Removing Response Not Clear from the Graph

Removing the NotClear responses, which is a small amount in proportion to the other responses, allows for the bar graph to scale larger and display a more obvious difference in the 3 primary types of responses. As I mentioned above, it is important for organizers and legislatures to pay attention to those who did not respond. They did not respond to the survey, but they may still vote on the matter and influence different results to this survey.

ausmarvotes %>%
  filter(Response == "Yes" | Response == "No" | Response == "NonResponse") %>%
  ggplot() +
  geom_bar(mapping = aes(x = Response, y = Total), stat = "identity") +
  labs(title = "Should Australian Law Change to Allow Same-Sex Marraige?", 
       subtitle = "2017 postal survey of Australians, excluding unclear answers", x = "Type of Response", y = "Number of Responses")

Concluding Thoughts

Something that I wish I was able to do was add labels on the boxplots in each bin to show the total count of responses. I was not able to figure out how to get this to work or how to adjust the way the y-axis number labels displayed… Another (more complex) bar graph that I think would be interesting is one that is shows the proportion of each type of response by City or by Division. You could use fill=Response to color code the type of response in each bar (City or Division) or you could use facet_wrap to display each Division as a subplot.

HW 5: Data Visualization

Megan Georges

11/23/2021