Page 105 - The World About Us

Padang Bai
1.1.1
1.1.1
3.2
2.2.3

The climate graphs on pages 10 and An observed correla on on a sca er Fig.274 Global palm oil produc on.
11 have no gaps between the blue bars. graph can o en be confirmed with a
This is because there are no gaps sta s cal opera on, such as performing
between the months of the year. Rain a Spearman's Rank calcula on. This will
falling at midnight at the end of the give a numerical value to a correla on, Indonesia 50%
31st January is s ll falling at midnight at backing up any observa on made using
the start of the 1st February. the line of best fit.
The mean monthly temperature is Sca er graphs are best made with a Malaysia 40.2%
plo ed in the centre of each monthly minimum of 10-12 data plots to allow a
column. The red line is drawn to extend clear trend to emerge. However, too Rest of the
from the start of the year to the end of many data plots could make the graph PNG 5.6% world 4.2%
the year, rather than mid-January to confusing. KEY
mid-December. Plo ng this way can be Dispersion graphs = 10% of global produc on
hard to do with many commonly used
database computer programmes. Dispersion graphs allow for an Pictograms
inves ga on of pa ern within the
Scatter graphs distribu on of data. Each value is Pictograms use symbols or pictures
Sca er graphs are excellent for plo ed on a ver cal axis. Two or more (usually simplified 'clipart' images)
presen ng data collected in pairs, for groups of data can be plo ed to allow displayed in propor on to the size of
example, river width and distance from for comparison. the data set being represented
the source. When combined with a line The dispersion graphs give a very (fig.274). Each symbol can represent
of best fit (which passes through the visual representa on of the spread of one or more of the items in the data, as
mean point of the data and is as close to data from the mean. They can iden fy long as that rela onship is made clear
the most of data plots as possible) the the similari es and differences in the in the key. While propor onal symbols
sca er graph is an excellent way of spread of data from two or more data are very visual, it can at mes be
iden fying correla ons. A rising line for sets. Figure 273 shows life expectancy difficult to read what exact quan ty a
le to right will indicate a posi ve data for ten randomly selected ACs, par al symbol represents.
correla on while a falling line from le EDCs and LIDCs. It shows that there is Pie charts
to right will indicate a nega ve greater spread of life expectancies in
correla on. LIDCs than ACs. In ACs life expectancies Pie charts are circles divided up
The closeness of the majority of are clustered towards to upper end of into segments to represent
data plots to the line of best fit will the data. In the EDCs the outlier is South percentages of a total. The percentage
figure for each group is mul plied by
indicate the strength of the correla on. Africa, s ll coping with the AIDS
Once again, the human brain is very pandemic. 3.6 to convert it into the degrees of a
circle. The conven on is to plot
good at iden fying these pa erns, as Dispersion graphs can be used to clockwise from a ver cal line (at 12
well as spo ng anomalies (points at plot the median value as well as the O'clock). The data for global
some distance from the line of best fit). upper and lower quar le and the inter- greenhouse gas emissions is presented
quar le range.
Fig.272 Sca er graphs and line of best fit. as a pie charts on page 79.
Fig.273 Selected life expectancies. Proportional circles
90 Propor onal circles are drawn in
y axes legend Posi ve correla on 80 propor on to the data that they are

represen ng. A common way of using
these is with the circles as pie charts.

the largest circle is used to determine
the size of all other circles. The same
x axes legend Life expectancy from birth, years 70 With propor onal circles the radius of
60
applies for propor onal pie charts.
Propor onal symbols are o en plo ed
y axes legend 50 onto a map to locate the data. On page
47 (figure 125) the data for the
employment structure of Indonesia is
40
displayed using pie charts. These have
Nega ve correla on
0 been scaled propor onally to represent
LIDCs EDCs ACs the total size of the workforce.
x axes legend
105

The World about Us.