Powell Ch 9 Analysis of Data
4 Roles of Statistics
#1 Indicate the central point around which a mass of data revolves
#2 Show the broad or diverse the spread can be for a mass of data
#3 Reveal how closely or distantly certain features within the mass of data are related
#4 Indicate the degree to which the facts might have occurred by mere chance, or if there is a probability of their having been influenced by some factor other than pure chance.
Basic Steps in Statistical Analysis
a) Categorizing the data
b) coding the data
c) calculating the appropriate statistics
Wildemuth Ch 29 Content Analysis
Content analysis - the systematic, objective, quantitative analysis of message characteristics. The meaning of the term message in content analysis is broad.
The analysis procedures should be unbiased, valid, reliable, and replicable.
Ch 30 Qualitative Analysis of Content
Qualitative content analysis involves a process designed to condense raw data into categories or themes based on valid inference and interpretation.
It is a valuable alternative to more traditional quantitative content analysis, when the researcher is working in an interpretive paradigm.
Ch 31 Discourse Analysis
Discourse analysis is a tool that can be used to uncover other meanings (aside from literal) - meanings that we negotiate in our everyday and professional interactions, but that are rarely made explicit within those interactions.
It offers ILS researchers the potential to explore paths of inquiry that can provide us with a far greater understanding of some of the phenomena that affect us all, but its practice takes rigor, clarity, and even self-inquiry.
Ch 32 Analytic Induction
It is a specific form of inductive reasoning (taking specific facts and arriving general conclusion) used to analyze qualitative data. It is a formalized method for developing and refining a theory or hypothesis, directly from the data.
It has not been used in ILS research at this time.
Ch 33 Descriptive Statistics
Descriptive statistics is to summarize your results.
Descriptive statistics are the most essential view of your study findings and are so critical components of any report of your research.
Ch 34 Frequencies, Cross-tabulations, and the Chi-square Statistic
Graphical displays of data should show the data clearly, make large data sets coherent, present many numbers in a small space, and avoid distorting what the data have to say.
It is worthwhile to report descriptive statistics on your study's findings, including frequency distributions on key categorical variables.
Ch 35 Analyzing Sequences of Events
It's not enough to know which behaviors/actions occur most frequently - we also need to know the sequence of those actions and which sequences occur most frequently.
You can use step-by-step approach through simple first-order Markov models and transition matrices, look for frequently occurring longer sequences using maximal repeating patterns analysis, or focus directly on the similarity between sequences through optimal matching algorithms.
Ch 36 Correlation
Correlation helps you to examine the relationship between two variables.
Specifically it is the proportion of the variability in one variable that is explained by the variability in the other variable.
Correlation can be graphically represented by a scatter diagram.
Ch 37 Comparing Means: t Tests and Analysis of Variance
Lots of research questions in ILS fields compare two systems, etc. You usually calculate a mean for each group. The problem is that because you probably are using a sample from your population of interest, you need to take into account the variability across different samples and within each sample. The t test and ANOVA can tell the likelihood that the difference you observed will hold tru across the entire population of interest.
Friday, October 21, 2011
Wednesday, October 12, 2011
Key Concepts: Methods/Ethics/Theory Readings
Powell Ch 3 Selecting the Research Method
Basic research is judged by its clarity of purpose and interpretation, by its ability to support or refute particular hypotheses, by the incisiveness of the new hypotheses it generates, by the generalizability of the results, and by its technical accuracy, but in addition by the degree to which the results can be utilized in developing a product, a process or a policy, to mention just a few types of application.
Applied research on the other hand, can validate theories and lead to the revision of theories. It 'takes theory and concepts from basic research and by formal methods of inquiry, investigates real world phenomena.'
The ethical researcher creates a mutually respectful, win-win relationship with the research population; this is a relationship in which subjects are pleased to participate candidly, and the community at large regards the conclusions as constructive.
Ethical practice guidelines for LIS professionals: stress that anonymity, or confidentiality, of everyone involved must be maintained; that library services should never appear to depend on patron participation in the study; and that no harm should come to any subject.
Powell Ch 4 Survey Research and Sampling
The basic assumption of survey research is that by carefully following certain scientific procedures, one can make inferences about a large group of elements by studying a relatively small number selected from the larger group.
The strength of survey research is in answering questions of fact and in assessing the distributions of the characteristics of populations.
Wildemuth Ch 6 Questions Related to Theory
Developing theory is by definition inductive - it moves from the particular to the general. Testing theory, on the other hand, is deductive.
You should plan to work with theories in the middle range - broad and abstract enough to be useful to the information and library science community generally, but concrete and well defined enough to be applicable to specific questions in the field.
Basic research is judged by its clarity of purpose and interpretation, by its ability to support or refute particular hypotheses, by the incisiveness of the new hypotheses it generates, by the generalizability of the results, and by its technical accuracy, but in addition by the degree to which the results can be utilized in developing a product, a process or a policy, to mention just a few types of application.
Applied research on the other hand, can validate theories and lead to the revision of theories. It 'takes theory and concepts from basic research and by formal methods of inquiry, investigates real world phenomena.'
The ethical researcher creates a mutually respectful, win-win relationship with the research population; this is a relationship in which subjects are pleased to participate candidly, and the community at large regards the conclusions as constructive.
Ethical practice guidelines for LIS professionals: stress that anonymity, or confidentiality, of everyone involved must be maintained; that library services should never appear to depend on patron participation in the study; and that no harm should come to any subject.
Powell Ch 4 Survey Research and Sampling
The basic assumption of survey research is that by carefully following certain scientific procedures, one can make inferences about a large group of elements by studying a relatively small number selected from the larger group.
The strength of survey research is in answering questions of fact and in assessing the distributions of the characteristics of populations.
Wildemuth Ch 6 Questions Related to Theory
Developing theory is by definition inductive - it moves from the particular to the general. Testing theory, on the other hand, is deductive.
You should plan to work with theories in the middle range - broad and abstract enough to be useful to the information and library science community generally, but concrete and well defined enough to be applicable to specific questions in the field.
Thursday, October 6, 2011
Article Review #2 Option B
David Riley 10.6.11
Stamatopols, A. & Mackoy, R. (July 1998). Effects of library instruction on university students' satisfactoin with the library: a lognitudinal study. College & Research Libraries. 69(4), 323-34.
Introduction
The purpose of this study was to determine how a library instruction program impacts patrons' overall, long-term satisfaction with a large urban university library. The goal was to define the 'drivers' of patron satisfaction at the library. My research topic is customer/patron service within an academic library - so determining the 'drivers' of patron satisfaction will be beneficial for my research.
Problem Statement
Stamatopols and Mackoy addressed the problem of how to know what drives long-term patron satisfaction. They focus on the relationship between a specific service, library instruction, and user satisfaction with the library.
Literature Review
The researchers spent a great deal of time in the article explaining the historical research in regards to how satisfaction has been evaluated, how library instruction has been evaluated, how objective measures have been used to evaluate library instruction, how subjective measures have been used to evaluate library instruction, how subjective & objective measures have been used to evaluate, and service quality versus satisfaction in library evaluation. Stamatopols and Mackoy studies builds off the historical research listed above to identify actual drivers of long-term patron satisfaction.
Method
The population used was composed of students in several section of an intro English composition course at a large urban university. The authors collected data three times during the semester: during the week preceding the library instruction (T1), during the week immediately following the library instruction(T2), and during the week the research assignment was due(T3), near the end of the semester. T1 questions focused on students' expectations of library collection, library staff, and computer & print based information, and basic demographic information. T2 questions were the same as T1 without the demographic questions. T3 questions asked about the actual experience and perceptions of library performance and their overall satisfaction with the library.
The authors focused on three questions with analyzing the data collected:
1) Which perceptions of performance are most/least associated with patron satisfaction?
2) Which perceptions of performance are or are not congruent with T2 expectations?
3) Which T2 expectations appear to have been influenced by the library instruction?
Analysis of the data was based primarily on two established procedures: the paried t-test difference of means test and correlation analysis.
Caveats
It would have been prudent for the authors to have asked questions about students' previous library experience as some of their results showed that the majority of students' expectations of how helpful the library staff were after working with them was left unchanged. This may have resulted from previous experience working with the staff.
Key Concepts: Data Collection Techniques Readings
Powell Ch. 5 Data Collection Techniques
Questionnaires and interviews are frequently used for obtaining information about a person's perceptions, beliefs, attitudes, and so on.
Observational methods are best suited for describing and understanding behavior as it occurs.
No one method is likely to be perfect for a given situation, but it should be possible to select one technique as the best alternative, given the objectives, subject, priorities, and limitations of the investigation.
Wildemuth Ch 18 Transaction Logs
Log data are used by researchers to study user-system interaction behaviors, such as search and browse behaviors, as well as system-mediated interpersonal interactions, such as reference and support chat services.
Trans log analysis is used to study many different types of systems, including online library catalogs, digital libraries, specific Web sites, and Web search engines
Ch 19 Think-aloud Protocols
Think-aloud protocols are a research method used to understand the subjects' cognitive processes based on their verbal reports of their thought during experiments. You request subjects to speak aloud, reporting what they are thinking while they are performing tasks during an experiment.
Compared to other data collection methods, the process of collecting think-aloud protocols is simple, but you will be able to collect valid data only when your study is designed well.
Ch 20 Direct Observation
The primary purpose of observation is to find out what people do; that is, this method of data collection is focused on the behaviors of participants in a particular setting.
Despite recent technological advances of data collection, direct observation will remain used as long as researchers are interested in what and how people think about experiences - seemingly indefinitely.
Ch 21 Participant Observation
In participant observation the researcher is a participant in the setting, such participation leads to a better understanding of the people and social processes that occur within that setting, and this understanding can lead to better theories about social processes in that setting and similar settings.
If this is done well it should appear as if the researcher is just hanging out with the subjects.
Ch 22 Research Diaries
Research diary can conceptually and physically resemble other data collection methods such as the questionnaire, the interview, or observation, but what makes it unique is that diary protocols require participants to make self-reports repeatedly over time.
There may be some limitations to diary studies (significant amount of time and commitment they can require from participants & researchers) careful design can mitigate these limitations.
Ch 23 Unstructured Interviews
In an unstructured interview the researcher comes to the interview with no predefined theoretical framework and thus no hypotheses and questions about the social realities under investigation; rather, the researcher has conversations with interviewees and generates questions in response to the interviewees' narration.
They are most useful when you want to gain an in-depth understanding of a particular phenomenon within a particular cultural context.
Ch 24 Semistructred Interview
Has predetermined questions, but the order can be modified based upon the interviewer's perception of what seems most appropriate.
One of the most useful data collection methods for studying a wide range of information behaviors.
Ch 25 Focus Groups
A group of individuals selected and assembled by researchers to discuss and comment on, from personal experience, the topic that is the subject of the research
One of the core strengths of focus group is that participants can compare their views with those of other participants in the group, rather than simply reporting their views to an interviewer.
Ch 26 Survey Research
The design of effective survey research involved "many decisions which need to fit together and support one another in a way that encourages most people to respond and minimizes inaccurate or inadequate answers".
Survey research design is dependent on careful planning that necessitates attention to a series of critical components to ensure effective data collection and implementation.
Ch 27 Measuring Cognitive and Affective Variables
Many ILS studies involve cognitive or affective variables that cannot be directly observed. In such situations, psychological assessment approaches can be used to measure cognitive or affective variables such as a person's attitudes, interests, beliefs, or feelings.
The most efficient way to ensure that our measures are reliable and valid is to use existing measures that have been implemented and evaluated in past studies.
Ch 28 Developing New Measures
While such research efforts are oriented toward methodological issues, rather than other types of research questions, they are of great importance to research and evaluation in ILS.
The main objective of measurement, especially measurement of cognitive and affective attributes, is to quantify the amount or level of a particular construct associated with each study participant.
Questionnaires and interviews are frequently used for obtaining information about a person's perceptions, beliefs, attitudes, and so on.
Observational methods are best suited for describing and understanding behavior as it occurs.
No one method is likely to be perfect for a given situation, but it should be possible to select one technique as the best alternative, given the objectives, subject, priorities, and limitations of the investigation.
Wildemuth Ch 18 Transaction Logs
Log data are used by researchers to study user-system interaction behaviors, such as search and browse behaviors, as well as system-mediated interpersonal interactions, such as reference and support chat services.
Trans log analysis is used to study many different types of systems, including online library catalogs, digital libraries, specific Web sites, and Web search engines
Ch 19 Think-aloud Protocols
Think-aloud protocols are a research method used to understand the subjects' cognitive processes based on their verbal reports of their thought during experiments. You request subjects to speak aloud, reporting what they are thinking while they are performing tasks during an experiment.
Compared to other data collection methods, the process of collecting think-aloud protocols is simple, but you will be able to collect valid data only when your study is designed well.
Ch 20 Direct Observation
The primary purpose of observation is to find out what people do; that is, this method of data collection is focused on the behaviors of participants in a particular setting.
Despite recent technological advances of data collection, direct observation will remain used as long as researchers are interested in what and how people think about experiences - seemingly indefinitely.
Ch 21 Participant Observation
In participant observation the researcher is a participant in the setting, such participation leads to a better understanding of the people and social processes that occur within that setting, and this understanding can lead to better theories about social processes in that setting and similar settings.
If this is done well it should appear as if the researcher is just hanging out with the subjects.
Ch 22 Research Diaries
Research diary can conceptually and physically resemble other data collection methods such as the questionnaire, the interview, or observation, but what makes it unique is that diary protocols require participants to make self-reports repeatedly over time.
There may be some limitations to diary studies (significant amount of time and commitment they can require from participants & researchers) careful design can mitigate these limitations.
Ch 23 Unstructured Interviews
In an unstructured interview the researcher comes to the interview with no predefined theoretical framework and thus no hypotheses and questions about the social realities under investigation; rather, the researcher has conversations with interviewees and generates questions in response to the interviewees' narration.
They are most useful when you want to gain an in-depth understanding of a particular phenomenon within a particular cultural context.
Ch 24 Semistructred Interview
Has predetermined questions, but the order can be modified based upon the interviewer's perception of what seems most appropriate.
One of the most useful data collection methods for studying a wide range of information behaviors.
Ch 25 Focus Groups
A group of individuals selected and assembled by researchers to discuss and comment on, from personal experience, the topic that is the subject of the research
One of the core strengths of focus group is that participants can compare their views with those of other participants in the group, rather than simply reporting their views to an interviewer.
Ch 26 Survey Research
The design of effective survey research involved "many decisions which need to fit together and support one another in a way that encourages most people to respond and minimizes inaccurate or inadequate answers".
Survey research design is dependent on careful planning that necessitates attention to a series of critical components to ensure effective data collection and implementation.
Ch 27 Measuring Cognitive and Affective Variables
Many ILS studies involve cognitive or affective variables that cannot be directly observed. In such situations, psychological assessment approaches can be used to measure cognitive or affective variables such as a person's attitudes, interests, beliefs, or feelings.
The most efficient way to ensure that our measures are reliable and valid is to use existing measures that have been implemented and evaluated in past studies.
Ch 28 Developing New Measures
While such research efforts are oriented toward methodological issues, rather than other types of research questions, they are of great importance to research and evaluation in ILS.
The main objective of measurement, especially measurement of cognitive and affective attributes, is to quantify the amount or level of a particular construct associated with each study participant.
Subscribe to:
Posts (Atom)