Use 11MP to teach ‘How to lie with statistics’

The subject of the 11th Malaysia Plan (11MP) is hard. It’s got to be told with numbers.

The 370-page report is beautifully produced. Many elements of good communication are present on every page.

The colour scheme is excellent. There is a lot of white space. There are few big words.

The paragraphs are short, but not repetitively so. Just the right number of font sizes are used.

There are a lot of graphics. Pages, which belong to the same chapter, are edged with the same colour.

There are many icons, perhaps too many. But this is a small fault. Overall, the presentation is excellent.

However, I am highly suspicious of the 11MP.

First, it’s hardly a plan, since it doesn’t say who will do what where and when. It’s more of a hope. (I wouldn’t go so far as to say it’s a vision.)

Second, it omits data which was provided in previous plans. I’ll illustrate this with two examples.

Example 1: In the 10MP, Table 10 projects that in 2015 the number of non-citizens in Malaysia will be 2.2 million, down from 2.5 million in 2010.

The 11MP omits the number of non-citizens, though Human Resources Minister Datuk Richard Riot said in November 2014 that the number exceeded 6.7 million. Why doesn’t the 11MP discuss this drastic failure?

Example 2: In the 10MP, Table 14 projects the number of enrolments in “Public and Private Institutions.” It projects that in 2015, for every 44 enrolments in public institutions there would be 55 enrolments in private institutions.

The 11MP omits the data for enrolments in the same set of institutions. It instead gives the ratio of enrolments in “universities and university colleges” for four years (2000, 2005, 2010 and 2013).

In 2013, the ratio was 54:46, which is the reverse of the projection for 2015. Why doesn’t the 11MP offer comparative data?

The above are two examples of evasion through omission of two contentious issues in Malaysia: (1) documented and undocumented foreign labour and (2) education.

The 11MP obscures the data. Why? Which reader will not ask “what else does the 11MP obscure?”

Many are wondering why the equivalent of “ownership of share capital (at par value) of limited companies by ethnic group” – found on page 403 in the 10MP – is missing from the 11MP.

Third, and perhaps most importantly, the 11MP uses graphics to deceive readers. The most glaring example is Exhibit 3-5: Target segments for Eleventh Plan (page 3-12).

Exhibit 3-5 is a visually attractive and stimulating graphic composed of 10 circles. The circles are of four different sizes, each in a different colour.

Each circle represents a different group. The groups are, from left to right: youth, family, B40 households, elderly, women, rural, Bumiputera, children, persons-with-disabilities and M40 households.

Yet Exhibit 3-5 is never mentioned in the text. This contravenes one of the basic rules of writing: never present an exhibit without discussion.

Since the 11MP scores so well in other aspects of writing, I am compelled to question whether this contravention reveals a deliberate attempt to deceive.

I’ll ask and answer the most obvious questions any conscientious reader would ask: “What rule do the authors suggest they used to decide the size of the circles?” and “Did they use the rule?”

The answer to the first question is “the 2014 data,” since the authors present it on the graphic. To answer the second question, I must evaluate the size of the circles:

(1) I observe that the largest circle – the one which draws the eye – is for the B40 (households with the lowest 40% of incomes), with 2.7 million households.

(2) I observe that there is another group, the M40 (households with incomes from the 41st to the 80th percentile), which also has 2.7 million households.

(3) I observe that there is an even larger group, the Bumiputera, whom the chart says comprise 68% of citizens (why don’t the authors indicate the number of Bumiputera households?); I calculate and figure this amounts to about 4.7 million households.

Based on the three things I’ve observed, the circles for the B40 and the M40 should be of equal size; and the circle for the Bumiputera should be much larger than the circles for the B40 and the M40.

However, the circle for the B40 is the largest, while the circles for the M40 and the Bumiputera are about the same size!

If different criteria were used to choose circle sizes, it is irresponsible to omit them from the chart.

The 11MP is intended to be debated in Parliament. Yet, it exhibits very strong evidence of evasion and omission.

It shows a blatant disregard for the Parliamentary process. It suggests transparency while obscuring data and analysis. I would use it to teach “How to lie with statistics”. – June 4, 2015.

* This is the personal opinion of the writer or publication and does not necessarily represent the views of The Malaysian Insider.