o Interesting parallels with field research Chapter 11 Analysis of Quantitative

# O interesting parallels with field research chapter

This preview shows page 9 - 11 out of 17 pages.

o Interesting parallels with field research Chapter 11: Analysis of Quantitative Data Quantifying Data o Based on closed-ended questions o Numbers represent values of “variables” which measure characteristics of subjects, respondents or cases o Typically gather information on many “variables” with well- established attributes (eg., gender with females coded as “1” and males coded as “0”) o Coding categories often establish beforehand, through modified with results in some instances o End product of coding = conversion of data to numerical codes to represent attributes of variables o Usually create a “codebook” that describes the coding procedures, variables and attributes and location of the data for analysis Creating Codebooks o Codebooks describe all actual questions/items coded, their locations in dataset, and what codes represent o Must be comprehensive and exhaustive Example: Country - Where does the sociologist work? Measurement level: nominal Column width: 8 Alignment: Right Value | Label 1.00 – Canada 2.00- United States 9.00 – NA/missing data o A codebook is required to define what each variable represents in a specific dataset, as well as the meaning of each attribute (datum) Cleaning Data o Refers to the process of detecting and correcting coding errors o Examine general fields to determine if data have been entered thoroughly, as well as whether there are missing data or other unusual responses o Possible Code Cleaning refers to process of determining whether only the valid codes assigned to variables appear in that column of the data file. (ex., Males = 0 and Females = 1 – gender should not have a 4) o Contingency Cleaning refers to determining if only particular cases have data entered that are specific to their situation (questions that apply to specific cases or subset). It involves cross-classifying two
variables and looking for logically impossible combinations. Ex., education is cross-classified by occupation. If a respondent is recorded as never finishing high school and is also recorded as being a lawyer o Univariate Analysis Stats used to summarize distributions of single variables in efficient ways Frequency Distributions is a table that shows the distribution of cases into the categories of one variable (the # or percent of cases in each category) Mode: most frequent occurring observation Median: observation that divides the distribution in half, after ranking all observations from lowest to highest, the observation at the exact centre of a distribution. (the middle point, if an odd # add the two middle scores together and divide by 2) Mean: arithmetic average of all observations (compute the mean by adding all scores together then dividing by the number of numbers) Most commonly employed measure of central tendency Range and Standard Deviation o Range : a measure of dispersion for one variable indicating the highest and lowest scores o Standard deviation evaluates average distances between all scores and

#### You've reached the end of your free preview.

Want to read all 17 pages?

• Fall '14
• JosephMichalski
• Sociology, Qualitative Research, researcher, Simple random sample