Textbook Notes (369,133)
Canada (162,403)
Sociology (1,513)
SOC244H1 (11)
Chapter

Ch 8 Quantitative social science.docx

12 Pages
127 Views

Department
Sociology
Course Code
SOC244H1
Professor
Cant Remember

This preview shows pages 1,2 and half of page 3. Sign up to view the full 12 pages of the document.
Description
Ch 8 Quantitative social science the surveyIntroductionquantitative research focuses on measuring quantities and relationships between attributes following a set of scientifically rigorous proceduresit collects highly structured data and is deductive in approach ie the investigator starts with ideas develops a theory and testable hypotheses from them and tests them with dataQuantitative research is approporaite in situations where there is preexisting knowledge about the phenomenon of interest which permits the use of standardized methods of data collection such the surveythe survey is the most common quantitative method which is used to describe social pehonomenathe survey is a method of collecting information from a sample of the population of interestthe unit of analysis in a survey is usually the individual although it can also be an organization ex Medical clinic or both of these in multilevel studieswhile most surveys are based on questioning people they can also take documents as their focus for research ex surveys of historical documents media outputs medical records or be based on direct observations ex As in traffic surveysdescriptive surveys are carried out in order to describe population ex Knowledge attitudes perceptions health behavior or other phenomena of interest to study associations between variables and to establish trendssurveys aim to collect information as accurately and precisely ass possibletheir distinguishing feature is that the same information is collected from each member of the survey samplethey can also cover large samples of ppla major advantage of surveys is that they are carried out in natural settings and random probability sampling if often easier to conduct that for experimental studiesa major advantage of surveys is that they are carried out in natural settings and random proabability sampling is often easier to conduct than for experimental studies This allows statistical ingerences to be made in realtion to the broader population of tinerest and thus allows generalizations to be madethis increases the external validity of the studysurveys which are carried out at one point in time are known as crosssectional or descriptive surveys and they aim to collect information about current nad past phenomena arid to explore assoications between variableslongituidnal surveys are conducted at more than one point in time either prospectively ex Surveying ppl at several points over time over the future course of time or retrospectively ex Going back in time analyzing records and aim to explore cause and effect relationshipsSampling for surveysSamplingin statiscal terms a population is an aggregate of people or objects Since the population of interest may contain too many members ex Ppl to study conveniently samples of the population from a complete and accurate listing of memebers are drawnsampling is cheaper in time staff and resourcesstatistical sampliong is recommended because when the estimates of the characteristcs of the population are calculated at the analysis stage the precision of the estimates can be determined from the resultssince all sample results are liable to be affected by campling errors the estimates should be accompanied by information about their precision This is known as the standard errorSampling errorall survey results are called estimates because they are subject to a range of errorssampling error is the probability that any one sample is not completely representative of the population from which it was drawnsampling error cannot be eliminated but it should reduced to an acceptable level The existence of sampling error means that whenever a hypothesis is tested there is a finit possibility of either rejecting a tru hypothesis type I error or accepting it when it is false type ii errorerrors can occur for 2 reasonssampling is not carried out properly resulting in a biased sample systematic errorthe other reason is the chance factors that influence the sampling process For example an unusually unrepresentative sample could be chosen random errorSampling framein order to be able to draw a sample a sampling frame is needed The sampling frame for a survey is the list of population members units from which the survey sample is drawnsurveys depend on it containing a complete and accurate listing of every element in the target population and every element should be included only onceinvestigaotros need to be aware of all the limitations of their sampling frames and attempt to address them in order to enhance the external validity of their resultsexternal validity relates to the generalizability of the research results to the wider population of intestestinternal validity refers to the psychomethtric properties of the measument instrumentsampling is concenered with sample selection in a manner that enhances the external generalizability of the resultsThe sampling unita member of the sample population ios known as a sampling unitthe sampling unit may be an individual a household an organization ex Hospital or a geogepahical areaSample sizeit is important to calculate the required sample size as accurately ass possible in order to be able to generalize the survey findings to the whole population of interestwhen the survey results are analyzed estimates of the population parameters are calculatedand statistical tests can be used to estimate differences in predefined outcome variables between groupsthe survey needs to be designed at the outset so that the sample size is sufficiently large to have a good chance of detecting significant differences between the group studied and where they existinvesitgators should also consider the need for subgroup analysis issues of item and total nonrepsonse and sample attrition in the case of longitudinal desings all of which will increase desired sample sizesthe probability that a statistical test will produce a difference between the groups tested at a given level of significance is called the power of the test
More Less
Unlock Document

Only pages 1,2 and half of page 3 are available for preview. Some parts have been intentionally blurred.

Unlock Document
You're Reading a Preview

Unlock to view full version

Unlock Document

Log In


OR

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


OR

By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.


Submit