How statistics are manipulated in everyday life
The book ” how to lie with statistics” is about how statistics are manipulated in everyday life to the advantage of some of people (Huff, 11). Several key ideas are pointing to this fact. One of the ideas is that it is almost impossible to obtain a perfectly random sample. A perfect sample is one which is large enough to be an adequate representation. It must also be random, which requires a lot of labor and resources to achieve. The next idea is that using non-random samples leads to bias. Biased samples can be avoided by using the strategy of stratified random sampling. Huff also believes that not all averages are created equal. This is created by the different types of averages such as mean, mode and median. Huff also indicates that marketers can use the chance to skew their results. They do this by using small samples which are statically insignificant. For example, they may report that 99 per cent of users of a certain product liked it, while the consumers of the advert have no idea what sample size was used (Huff, 45). The other idea is that readers should be cautious of the missing standard error. Statistics can also be “too good to be true” (Huff, 100). One can find out the legitimacy of statistics by finding the standard error, given that the deviation is known. Huff also advises readers to be careful not to fall for arbitrary comparisons. This is when someone wants to make a conviction that something is true, but they don’t have the proof. They instead use something else and assume it’s the same. On the other hand, not all errors have a bad intention. Some result from misunderstandings, and the fact that there are different correlations. Huff, however, advises readers to always be doubtful.
Real-life example
The human mind is controlled by 2 systems which are constantly fighting over control of a person’s behavior and actions. Reading “How to lie in statistics” would help someone to learn the different ways in which this would result to errors in memory, judgment and decisions, and what they can do about it (Huff, 123). This is especially while making a decision about money where emotions come in and prevent one from making a concrete decision.
For example, there is a scenario where two neighbors are having a conversation. One of the neighbors A is of the idea that they cannot pay someone to come and mow their lawn with $20 because this would be overpriced. Neighbor B offers A the same amount of money $20 to come and mow their lawn but A refuses.
This indicates that the neighbor has a problem with understanding statistics, and rather they use their first system to make impulsive decisions related to emotion and fear of losing money. If they can study the book on how to lie with statistics they can learn not to jump into conclusions first, be aware of the arbitrary comparisons and use of facts and statistics to come into a conclusion (Huff, 139).
The neighbor fails in the argument because they did not try to consider statistics, probability like what it takes to mow a lawn. They don’t take the time to make a sample data in their mind of the costs that would be involved in the activity and when the odds are in their favor, act accordingly.
If they read the book “How to lie with statistics”. They would be able to do a statistical analysis of the work needed to be done and the cost implication. After getting all the information then they can decide to either manipulate the worker to lower the costs with reasons, accept the offer or do it themselves.
Managerial application
This book contains quality lessons for carrying out managerial activities. Any manager will have several instances where people will approach him with proposals to do business with them. In most business proposals, statistics are applied to show how a certain product has been performing or how it is expected to perform in the future. The book has explained how to determine whether these statistics are legitimate. One of the key things is to enquire on what sample size has been used in attaining such results. As Huff explains, this should be large enough to avoid bias and misrepresentation (Huff, 69). Besides this, one should also look for bias regarding how the sample was selected. For example, some samples may be included or omitted from the study depending on how they affect they influence it. Where averages mentioned have a significant effect on the statistics, one should be keen as to which type of average is used, since people may choose to apply the one that favors their proposal (Huff, 102). As a manager, such points should also be kept in mind when making such statistics for one’s own company. For example, when carrying out a market study on the products of the company, all the proper rules of statistics such be applied to get accurate results which will steer the company ahead.