Search This Blog

Friday, April 14, 2017

Unit 7 - Data Analysis Techniques / Connaway & Powell, Chapter 9

Analysis of Data
"A knowledge of basic statistics is imperative to research producer and consumer in library science
statistical analysis is concerned with development and application of methods and techniques for organizing and analyzing data quantitatively so that the reliability of conclusions based on the data may be evaluated objectively in terms of probability" – page 261

Types of statistics – theoretical/applied

  • theoretical equal mathematical aspects of statistics
  • applied involves the practical application of statistics


Role of statistics – four basic purposes
1. – statistics show central point around which a mass of data revolves
2. – show how broad or diverse the spread can be in a mass of data
3. – reveal how closely or distantly certain features within the mass of data are related/relationship?
4. – degree to which the facts have been incurred by mere chance/probability question mark influenced by some factor other than pure chance

Cautions – cannot interpret
necessary for studies involving quantitative data/crucial at the sampling and analysis stage
study should not be dictated by the statistical techniques to follow

Steps involved in statistical analysis
establishment of categories take place before the data are gathered

four basic rules or guidelines
1. set of categories or values – derived from a single classificatory principle
2. each set of category should be exhaustive – possible to place every observation in one of categories of the set
3. each set category should be mutually exclusive – not possible to place a specific observation correctly and more than one category
4. based on sound knowledge of the site of subject matter and anticipation of likely responses

Coding the data
categories have been established and data assigned to them – necessary to convert the new data responses to numerical codes – tabulated or tallied
original responses numerical? – Do not need to be assigned new codes

Important consideration: Reliability / poorly worded question near item may not produce enough relevant information for the purposes of the study
 asked more than one question
problems with the categories may lead to lack of reliability
important that coders are adequately trained and important to verify or check the accuracy of their work

researchers transcribed observations or scores for each individual or case
from the data collection to coding or transfer sheets
analyze the data manually
if large mass of data/statistical techniques to be employed were relatively complex and time-consuming to conduct – use a computer to analyze the data


  • Optical scanning for producing data flow files
  • direct data entry may not be the most common technique for creating data files
  • computer program displays each question on the screen
  • prompts the researcher to input the response directly on the screen – and into the computer file
  • database management program to control data entry
  • statistical programs now enable downloading a computer datafile rent tearing data directly into the program using the data editor
  • program prompts the person entry data for each response
  • checks the response to ensure its valid
  • phase response of the property to file


After data has been prepared and saved to electronic file – analyzed using a software program

Analyzing the data date – descriptive statistics
descriptive statistics – predominant type of data analysis employed by researchers in library and information science
six basic functions
1. indicate how many persons objects scores or whatever achieved each value – frequency distributions/reported in tables
2.  difficult to grasp the overall meaning of a frequency distribution tables? Pictorial representations – pretrip variety of characteristics of cases or individuals
indent paragraphs charts pie charts histograms frequency polygons
graphs are useful with a large number of cases
3.  – typical – recent measures of central tendency
mean median mode
4.  how widely the cases in a group very
dispersion or variability
mean aviation – arithmetic mean of the absolute difference of each score from that mean
standard deviation – square root of the arithmetic mean of the square deviations from the mean
variance – the mean squared aviation
standard deviation is one of the most frequently used measures of dispersion – most difficult to comprehend
5.  measure the relationship – correlation or associational statistics
cross tabulation or by via direct frequency – products of tables in which two variables have been cross classified – table consists of rows and columns of one variable labels for the rows of another variable
finally compare percentage differences
6.  basic function – descriptive statistics can perform is to describe the difference between two or more groups of individuals
special case – showing the relationship between variables
measures of central tendency

Analyzing the data – inferential statistics

No comments:

Post a Comment