雅思阅读第100套P2-Sampling Bias
Reading Passage 2
You should spend about 20minutes on Questions 14-26, which are based on Reading Passage Two.
Sampling Bias
Our primitive ancestors leftmany paintings on the walls inside caves. Additionally, inside and near theseplaces there is evidence of fire pits, and refuse and burial sites. However,one could equally imagine this same evidence of daily life on exposed cliffs orhillsides, on trees or animals skins, and beside rivers and coastlines. Suchevidence, if it existed, would have long been washed, eroded, or rotted away.Thus, prehistoric people are characterised as 'cavemen’, presumed to have apredilection for dwelling in these places only because that is where mostevidence is taken. This 'caveman effect’ is an example of what is known as'sampling bias’ — one of the biggest problems when conducting any form ofstatistical data gathering.
Surveys, for example, arepopular because they are easy to administer and relatively cost-effective,particularly if conducted remotely through technical means, such as telephone,mail, email, or the Internet. Surveys also lend themselves to obtainingparticularly large numbers of respondents, which, in theory, allows a greaterchance of sampling all the variations of the target population. They can alsobe standardised with fixed questions and responses (such as 'tick the box’ or'closed-ended' questions). This allows easy collation, analysis, andpresentation of results, all with the air of precision that mathematics brings.Such surveys, however, have proven notoriously unreliable because of thedifficulty in obtaining representative samples. In other words, the sampling isbiased, or skewed in favour of certain outcomes.
Let us look at some examples.If one calls people on cellphones, it immediately excludes those who favourlandlines, and thus the sample of respondents may be those who are moretechnically-conversant, skewing data based on, say, technical issues ('Howoften do you use the Internet?’). If one rings domestic homes during thedaytime, most of those who work during the day will be excluded. Those thatanswer will more likely be the unemployed, disabled, elderly, and retired,skewing data based on, say, work-related issues ('How important is work in yourlife?’). No matter how large the sampling size is, sampling bias canimmediately invalidate the results.
One of the more subtle ofsampling biases is known as self-selection. No matter how rigorously therespondents are chosen to be random and characteristic of the targetpopulation, those who choose to respond will be different to those who do not.Generally, respondents who are willing to invest time in giving answersobviously want to say something, whereas those who choose not to answerprobably do not. Thus, any survey in which many respondents do not answer, donot give clear answers, or only give cursory or unthinking answers, isimmediately invalidated, since opinionated perspectives are disproportionatelyrepresented.
The latter is such an immediateand obvious problem that it has given rise to techniques to maximise thepossibility of garnering responses. One of the more effective is to give therespondents advanced warning (often through the mail), highlighting the time,the nature of the survey, and the mode of delivery, as well as expressingappreciation for the assistance. The interviewers themselves must besufficiently trained in correct question-asking techniques, and, with cranks,salespeople, and scam-artists abounding, interviewers must provide introductionsabout themselves, their company, and the nature of the interview, fully andwith evident sincerity, in order to gain the trust of those they are talkingto.
Even with this, sampling biascan easily arise due to the number of variables in place, since it only takesone to skew the data. If taking samples from a specific location — say, astreet corner—then it may be that this location is in the business district,excluding ordinary workers from the sample. It may be that it is near arestaurant district, excluding those who cook more often for themselves. Ifthere is a health club nearby, the majority of respondents may be muchhealthier than the average of the population. If it is on a university campus,designed to poll university students, is it near the engineering or the artsfaculty? The part-time or full-time schools? Are they rich or poor? Male orfemale? What about race, colour, gender, religion, socio-economic background,and first language? The list goes on and on.
One method to deal with this isto make sure all targeted groups are represented, if only a little, and makemathematical extrapolations to correct the bias. For this to work, the degreeof underrepresentation needs to be quantified exactly, and one needs to assumethe under-represented respondents are indeed typical of their kind. If, forexample, one aims to find the opinion of the population regarding the outcomeof an election, but could only, for whatever reasons, interview one woman forevery four men, the responses of the women could be multiplied by four, andthus, one can assume (guardedly and with many provisos), that the sampling biasfrom gender has been corrected. But that does assume all the other variableswhich introduce bias have been excluded — often a very problematic assumptionto make.
SECTION 2: QUESTIONS 14-26
Questions 14-18
Do the following statementsagree with the information given in Reading Passage Two?
Write:
TRUE
if the statement agrees with the information
FALSE
if the statement contradicts the information
NOT GIVEN
If there is no information on this
14 _________________ Cavemen were often very good artists.
15 _________________ Surveys can be done cheaply by telephone.
16 _________________ Surveys can usually give reliable information.
17 _________________ The elderly and disabled people are often athome during the day.
18 _________________ Larger survey samples can reduce sampling bias.
Questions 19-24
Complete the flowchart.
Choose NO MORE THAN TWOWORDS from the passage for each answer.
19 _________________
20 _________________
21 _________________
22 _________________
23 _________________
24 _________________
Questions 25-26
Choose the correct letter,A, B, C, or D.
25The number of sampling variables
Ais usually not so large.
Bcan result in important input being lost.
Cmeans many locations need to be used.
Dcan result in lists being necessary.
26Mathematical extrapolation
Acan yield confident results.
Brequires responses from both men andwomen.
Cneeds exact ratios.
Dneeds many respondents.
做真题，得高分，好录取。
托福真题网：
http://www.tuonindefu.com答案＋we-chat : 836064952
答案
雅思阅读第100套P2-Sampling Biashttp://www.tuonindefu.com/?p=2746雅思阅读第100套P2:Sampling Bias
