bodyType

customContents

entityId

entityUuid

articlePageQuery

filter

article

articleBaseAdvertisingInfoArticle

articleListArticle

articleSeoArticle

helpersCheckIsPostiesArticle

hooksArticleAcknowledgementGateOverlayArticle

hooksTrackPageViewArticle

hooksContentInterestArticle

articleLinkingDataContent

hooksAmpRedirectArticle

articleSchemaContent

liveArticle

advertZone

sentiment

contentLock

topics

types

sections

paywallTypes

createdDate

publishedDate

moreOnThisArticles

urlAlias

commentCount

authorLocations

authors

corrections

sponsorType

displaySlideShow

multimediaEmbed

printHeadline

seriesContainer

articleSpeeches

socialHeadline

headline

images

flag

firstTopic

glossary

flattenTypeIds

additionalInfoBoxes

quizIds

image

relatedLinks

relatedNewsletters

__typename

sponsor

pdfFiles

seriesReverseOrder

series

disableOffPlatformContent

published

updatedDate

summary

subHeadline

body

keywords

readingTime

customTimezone

liveContents

SCMP Reporter

Gathering of facts is a science, so it's important to get the building blocks right Statistics are important because they help us understand society and tackle social and economic issues in an objective, scientific manner.

  Apart from statistics published by government departments, you also come across data on specific topics prepared by research companies or other organisations.

  A basic knowledge of proper statistical methods enables us to differentiate between reliable statistics and those which have not been compiled professionally.

  When presented with statistical information, we should always find out what is the source of the data - that is, the 'raw' information used in the compilation of statistics. Raw data should be distinguished from actual 'statistics'.

  It need not be numerical - for example, sex, occupation, and so on.

   Basically, statistics are the information derived from - or summarised from - the raw matrial.

 Data on the income of individual households is raw, whereas data on the average household income derived from this raw information can be called statistics.

  There are three main sources of raw data: administrative records, special returns and surveys.

  A government (or any organisation) keeps administrative records for specific purposes - from these, useful statistics may be compiled.

  For instance, statistics on births, deaths and migration can be compiled from the government's birth and death registration records and from immigration clearance records.

  The statistics compiled are actually by-products of such administrative systems.

  In the private sector, sales records are often the source of sales statistics, which are very important for planning marketing strategies.

  Special returns refer to those returns which are collected primarily for the purpose of compiling statistics.

  A typical example is the 'trade declaration' which the Government requires every trader to complete when he or she imports or exports goods.

  Based on the information in trade declarations - country/ territory of origin and destination, value and description of goods - detailed statistics on Hong Kong's trade with other countries are compiled.

  In a survey, the objects of study are interviewed, observed, or given questionnaires to fill in.

  This is the most widely used means - in both the government and the private sector - for collecting information for the compilation of statistics.

  Before carrying out a survey, we have to be clear about the 'target objects' for data collection. These objects could be individuals or firms; they are known as 'elements' in a survey, with the overall aggregate of such elements called the 'population'.

  Normally, an element is also the survey respondent.

  Occasionally, respondents are not the survey objects.

  Rather, they provide information on behalf of others. A survey in which each and every element of a 'population' is enumerated is known as a census. A well-known example is the population census conducted in Hong Kong every 10 years.

  Where only a small grouping of elements is taken, it is more clearly described as a 'sample survey'.

 The Government regularly conducts a number of sample surveys with the aim of getting information on various economic or social aspects of Hong Kong.

   Some examples are statistics on marriages, job vacancies, wage rates, sales and production.

 There is a whole body of knowledge on the subject of sampling, with much scientific basis.

  Haphazardly chosen samples - those comprising elements selected at will in a convenient fashion - may not generate valid results.

  'Random sampling' refers to certain kinds of scientific sampling and not sampling which is has been carried out haphazardly.

For more information on this series of articles, please write to the General Statistics Branch (2) of the Census and Statistics Department at Wanchai Tower, 12 Harbour Road, Wan Chai, Hong Kong, or call 2582-4004 



Raw data fuels surveys