LIS 504 - Terms with examples


|A |B |C |D |E |F |G |H |I |J |K |L |M |N |O |P |Q |R |S |T |U |V |W |X |Y |Z

A

action research

"This paper discusses a course partnership involving a large engineering and professional services company, and a public university, both headquartered in Philadelphia. An action research study of the course partnership...is used as a basis. ... Like typical action research studies..., ours aimed at providing a service to the research clients...while at the same time performing an exploratory investigation of the effect of Web-based collaboration technologies on course partnerships." (Kock, Auspitz, and King, 2000, p. 158)

alternative hypothesis

"The alternative hypotheses for Hypotheses H0 (1) to H0 (4) were that there was a significant difference in mental models among and between different groups of subjects..." (Zhang and Chignall, 2001, p. 448)

anonymity

"First, in order to insure anonymity, each questionnaire was identified only in terms of the city and the groups to which each participant belonged. By mailing the postcard separately, the anonymity of the returned questionnaire was secured and, for follow-up purposes, the respondents were informing the researcher that they had mailed the questionnaire." (Mittermeyer, 1999, p. 11)

"Respondents were able to complete the questionnaire in less than 10 minutes and their anonymity was guaranteed." (Speier, Palmer, Wren, and Hahn, 1999, p. 539)

association

"...scholars with a higher self-perceived ability to use the Internet tended to use e-sources more often." (Zhang, 2001, p. 648)

"Correlation analysis was performed to see if there were any tendencies for certain kinds of citations to be associated with each other (i.e., books with journals, and websites with magazines)." (Davis and Cohen, 2001, p. 311)

availability sampling

"A convenience sample of 15 participants took part in two separately scheduled sessions." (Gottlieb and Dilevko, 2001, p. 520)

"These topics were supplemented by selecting five queries from another study." (Leighton and Srivastava, 1999, p. 873)

average

"An average of 1.55 attributes affected each of the classificatory decisions for bookmarks that were evaluated within the 15 participant questionnaires..." (Gottlieb and Dilevko, 2001, p. 526)

"Table 2 gives the descriptive statistics for the sample of Ph.D. (or M.D.) respondents. The average age is 49." (Walsh, Kucker, Maloney, and Gabbay, 2000, p. 1297)


B

bias

"As Van House and Childers acknowledge, ... their sampling process (relying on library directors to distribute the questionnaires) may have biased the sample towards similarity of perception..." (Mittermeyer, 1999, p. 7)

"While the ISI databases contain citations to published items of any sort (including books and book chapters), they draw citations only from a set of several thousand carefully selected journals... Consequently, citation identities obtained by Dialog searches in the ISI databases will exclude data from what may well be an author's major works." (White, 2001, p. 91)

bibliometrics

"Bibliometrics... has been used extensively for studying the publication record of research faculty and their departments, and also to understand how scholars communicate through their publications. There has been very little application of bibliometric research to undergraduate research papers... Studying the bibliographies of student term papers can provide an account of the resources consulted in the research process, and a measure by which to evaluate undergraduate scholarship." (Davis and Cohen, 2001, p. 310)

C

case study

"The eight authors selected as test cases in this study are all from information science... For each, I shall exhibit what I call a citation identity... I use information scientists with long publication records because readers of this journal are likely to know something of their writings..." (White, 2001, p. 88)

"Unfortunately, little case study material exists on the development of network technologies in the public sector... The authors of this article hope that the following case study will be a step in meeting this need by examining the Federal Management Agency's (FEMA's) evolutionary relationship with information technologies." (Ward, Wamsley, Schroeder, and Robins, 2000, p. 1019)

causation

"Several early studies found that, because CMC [computer-mediated communication] filtered out cues that support a sense of 'social presence,' such as facial expressions and body language, it tended to generate increased antisocial behaviors among participants..." (Burnett, Besant, and Chatman, 2001, p. 539)

"On intuitive grounds it may be assumed that barriers to knowledge precede and partly create barriers to behavior, in that not knowing something seems a logical reason for not doing it. But very likely the process works the opposite way as well, if barriers to behavior ('don't discuss this topic') in turn create barriers to knowledge, as people are inhibited from sharing any information that they do possess." (Sligo and Jameson, 2000, p. 864)

chi-square

"Chi-square tests were performed to identify differences between types (or categories) of references cited in 1996 and 1999." (Davis and Cohen, 2001, p. 311)

closed question

How would you rate your overall ability to use Internet? (Please circle ONE number closest to your level.)
 No use at all .......... 0
 Beginner .......... 1
 Below average .......... 2
 Average .......... 3
 Above average .......... 4
 Expert .......... 5
(Zhang, 2001, p. 652)

How well do you identify with the following statement? If the library has what I need I would be able to find it using the online public access system.
[] strongly agree
[] agree somewhat
[] I do not agree nor disagree with this statement
[] disagree somewhat
[] strongly disagree
(Slone, 2000, p. 768)

cluster sampling

"The cluster-quota sampling technique with equal allocation method was used for sample selection. A sampling fraction of 72.2 percent was adopted for selecting clusters of 26 states out of a total of 36 states in Nigeria. From the selected states, ten consumers/patients of health service each were drawn giv[ing] a total sample size of 260." (Popoola, 2000)

concept

"...Wurman's concept of the information anxiety that covers both cognitive overload and disorientation problems..." (Heo and Hirtle, 2001, p. 670)

"Four concepts constitute the theory of normative behavior: ... (2) wordview ... Worldview is a collective perception held in common by members of a social world regarding those things that are deemed important or trivial." (Burnett, Besant, and Chatman, 2001, p. 537)

confidence interval

Appendix C: Median, Mean, and Mode for Dead, Duplicate and Zero Links
Query\Serv Alta V Excite HotBot Infoseek Lycos
Dead median 1 0 1 3[2..5]* 4.5[3..6]*
...
*Used an estimated population mean and a confidence interval for the median based on the the Wilcoxon test.
(Leighton and Srivastava, 1999, p. 881)

confidentiality

"...67 term papers from 1996 and 69 term papers from 1999 were collected from the professor. Bibliographies were photocopied and numbered for analysis, and any personal information was removed to preserve confidentiality." (Davis and Cohen, 2001, p. 310)

"The study took place at a Middle School (named Middle School for confidentiality purposes) located in East Tennessee." (Bilal, 2001, p. 122)

content analysis

"To analyze the qualitative data [from those] interviewed, content analysis was employed. One of the primary objectives of this content analysis was to develop a typology of scholars' motivations for hyperlinking in scholarly electronic environments. Therefore, the unit of analysis was a 'reason' why an author used a hyperlink in his or her scholarly e- article." (Kim, 2000, p. 889)

"The interviews were transcribed from the tapes and content analysis was used to analyze the transcripts. The study used three steps in data analysis, as suggested by Strauss... First, each participant's transcript was examined to assess whether all of the major research issues were included. Second, all transcripts were reviewed to determine the most significant participant perspectives on each issue. Third, pauses were taken during the coding to record ideas, concepts, or categories triggered by the analysis." (Sligo and Jameson, 2000, p. 862)

control group

"To test the effectiveness of visual tools in general, a control group for each Web space setting was included in the experiment." (Heo and Hirtle, 2001, p. 668)

"Four conditions were tested: Detection, Pension Calculation, Detection with False Alarm Penalty, and Control. ... The $100 prizes were not contingent on task performance in the Control condition." (Klein, 2000, p. 100)

cross population generalizability

"Since the study participants were all residents of New Zealand, there may be cultural peculiarities that impact upon this data. The dominent New Zealand culture is much more similar than different to mainstream North American culture..." (Julien and Michels, 2000, p. 14-15)

"...this study of one undergraduate class at Cornell University cannot be considered representative of college students in general..." (Davis and Cohen, 2001, p. 312)


D

definition

"For the purpose of this study, the term Internet-based electronic resources or e-sources was broadly defined to include sources that are available via Web browsers, FTP, gopher, telnet, mailing lists, e-mail, or other network tools or protocols, such as the following: ...". (Zhang, 2001, p. 629)

"Information behavior can be defined as a state in which one may or may not act on available or offered information." (Burnett, Besant, and Chatman, 2001, p. 537)

dependent variable

"The dependent variable, frequency of use, ...was converted to a common and continuous measure - how many times scholars used e-sources in a month." (Zhang, 2001, p. 636)

"The two dependent variables of performance (accuracy of the answers and time accomplish tasks in seconds) in each test (pretest and main test) were extracted from the log file and measured." (Heo and Hirtle, 2001, p. 670)

descriptive statistics

"For the purpose of this study, both inferential and descriptive statistics were used. ... The highest percentage of the respondents are farmers and civil servants while the lowest proportion of them are traders. Also, of the 250 respondents, 119 (47.6 percent) are males and the remaining 131 (52.4 percent) are females...." (Popoola, 2000)

distribution

Twenty-five Most Frequently Occurring Term Pairs [in Excite queries]
Term Term Frequency Term Term Frequency Co-occurrence frequency
and 19974 and 19974 6116
of 10433 the 6512 1901
... ... ... ... ...
(Wolfram, 1999, 33)

Sample Distribution. Group 1. City Council Members
1 City Council with 19 members
2 City Councils with 13 members
5 City Councils with 11 members
...
(Mittermeyer, 1999, p. 8)

E

ecological fallacy

"...the well-known inverse relationship between Recall and Precision does not allow one to predict what the precision or recall of a particular retrieval experiment will be, and the Bradford-Zipf 'law' does not allow one to predict the size of a core corpus of literature in a particular subject." (Holmes and Oppenheim, 2001)

empirical research

"Empirical studies of people's relevance criteria have investigated a great number of factors, which people use when they make their judgments of document relevance. ... The work reported here involves two sets of empirical investigations of people's perceptions of the importance of criteria and their actual use of criteria during real-time episodes of document evaluation." (Tang and Solomon, 2001, p. 676)

ethnography

"...we present three edited selections of the lived work occurring in the library. ...we hope that these snippets serve to elucidate the ethnomethodological approach to discover and analysis in practical detail." (Crabtree, Nichols, O'Brien, Rouncefield, and Twidale, 2000, p. 673)

ex post facto hypothesizing

"Because we use the Geissler criteria in part to reconstruct the Geissler list, then test our method by measuring how well it succeeds in producing the list, the issue of apparent circularity should be addressed." (Swanson, Smalheiser, and Bookstein, 2001, p. 799)

experiment

"Two closed Web environments were constructed for this experiment by extracting two sets of existing Web pages at the University of Pittsburgh that were likely to be previously unknown to the participants. For a large Web space, a set of 583 pages ... For a small Web space, a set of 50 pages ..." (Heo and Hirtle, 2001, p. 669)

"The experiment... Given an assignment to prepare an outline for a presentation on 'Year 2000 Problem and Its Social Effect,' participants were instructed to choose articles that would have the information necessary to them to produce this outline ..." (Tang and Solomon, 2001, p. 680)

experimental design

"...there was only one instance of community filtering... Therefore, the effect of different community filterings was not tested, and a symmetrical experimental design is not warranted." (Shepherd, Watters, Duffy, Gugle, 2001, p. 152)

"All users performed the same retrieval task... under the three experimental conditions. This experimental design is called 'repeated measures design'..., and the order in which users perform the tasks may influence their performance... Therefore, half the users performed first the task under condition V, whereas the other half performed first the task under condition T. ...It was decided that all subjects should perform the task under condition C last..." (Tombros and Crestani, 2000, p. 935)

experimental group

"In the Detection and Detection with False Alarm Penalty conditions, subjects were told that their performance on the task would be evaluated...and that the thirty percent of subjects with the best performance would be entered into a lottery.... from which one person would receive $100. The $100 prizes were not contingent on task performance in the Control condition." (Klein, 2000, p. 100)

explanation

"...the breakdown of Lotka's law... coincides with a change in the behavior pattern of the scientists... At that time high energy physics in The Netherlands suffered a decrease in funds...leading to difficult research conditions. These, in turn, led to a modification of the social structure among scientists, as shown by their collaboration patterns." (Kretschmer and Rousseau, 2001, p. 612).

"There are a number of potential explanations as to the preponderance of Content Attributes observed in the analysis of participants' classificatory decisions for bookmarks. From a methodological standpoint, the fact that the participants were working with a pre-selected set of URLs, rather than their own bookmarks, and the wide range of web sites used in the study may be contributing factors. If the web sites utilized in the study were not previously familiar to the participants, the Contextual Atrributes Use and Knowledge/Interest would not be particularly applicable." (Gottlieb and Dilevko, 2001, p. 527)

external validity

"In general, relative recall results are consistent with previous recall analyses, which indicate that automatic QE via semantic relationships generally improves recall... although not always significantly..." (Greenberg, 2001, p. 409)

"The small sample size (17) may impact the external validity of the results and their generalization to the whole population of seventh-grade children." (Bilal, 2001, p. 123)


F

face validity

A claim for high content validity was made on the basis that:
(Schamber, 2000, p. 743)

fallacy

"...what may appear on the surface to be an individual interest barrier, after fuller assessment, may be more accurately described as a collective or external cultural barrier." (Sligo and Jameson, 2000, p. 863)

field research

"...it is common to see ethnographic approaches employed in LIS reserarch, approaches marked by the immersion of a fieldworker within a library setting, gaining an insider's familiarity with, and gathering descriptions of, given areas of activity." (Crabtree, Nichols, O'Brien, Rouncefield, and Twidale, 2000, p. 667)

filter question

Are you satisfied with the current state of e-sources as sources for your research?
 Yes .......... 1 -> (SKIP TO Q.13)
 No .......... 2
(Zhang, 2001, p. 654)

focus group

"...it was decided that using a focus group approach would not be culturally acceptable, as this would involve the women talking in front of their peers about a highly sensitive topic..." (Sligo and Jameson, 2000, p. 862)

"After reviewing the design literature with a focus on participatory design and user-centered design, the best match between this project's needs and the user groups was determined to be a focus group methodology employing flexible mock-ups. ... A focus group approach provided a means for bringing the users and the technology together; we could demonstrate the technology and let the users experiment with it if they so desired. ... A focus group approach allowed users to come together for a concentrated period of time to provide design insights. It also allowed the users to debate among themselves, highlighting differing perspectives on the technology and its design." (Vaughan and Schwartz, 1999, p. 589)


G

generalizability

"The children who participated in this study may not represent all middle-school students in Tennessee, nor may they represent the general population of seventh-grade science students." (Bilal, 2001, p. 123)

"...the generalizability of the findings of the qualitative interviews would be enhanced by the supporting results of the subsequent quantitative mail questionnaire survey research..." (Kim, 2000, p. 891)

grounded theory research

"From the sample, we developed, tested, and applied a classification scheme, using a grounded theory approach..." (Spink, Wolfram, Jansen, and Saracevic, 2001, p. 233)

H

histogram

"FIG. 1. User-perceived term associations between all the terms identified in the rank list and the query terms.", "FIG. 2. Term relationships between the five best terms selected by users and the query terms." (Efthimiadis, 2000, p. 995-996)

historical research

(Flood, 2000)

(Ward, Wamsley, Schroeder, and Robins, 2000)

hypothesis

"Automatic QE [query expansion] via synonyms and partial synonyms (SYNs) will retrieve additional relevant documents not retrieved via the basic thesaurus search and increase relative recall. SYNs for a given set of search terms, searched together as a group, will increase relative recall with a loss of precision that is not significant." (Greenberg, 2001, p. 404)

"H1 - users will prefer community filtering to a blend of community and personal filtering." (Shepherd, Duffy, Watters, and Gugle, 2001, p. 152)


I

idiographic

"The fall of the Soviet Union, and the subsequent drop in the threat of a nuclear war, undermined the agency's primary mission, civil defense." (Ward, Wamsley, Schroeder, and Robins, 2000, p. 1029)

"A new set of personal relationships that benefitted the emerging decentralized ERIC can be traced back to the election of President Kennedy." (Burchinal, 2000, p. 570)

independent variable

"The examined independent variables included age, gender, number of access points to various Internet tools and resources, number of years of Internet experience, self-perceived overall ability to use the Internet, and satisfaction with e-sources. The main question of interest was the prognostic significance of these explanatory variables, that is, which, if any, of these variables were predictive of how frequently the respondents used e- sources." (Zhang, 2001, p. 636)

"The science teacher rated children's domain knowledge (i.e. science), topic knowledge (i.e., ozone), and reading ability on a 10-point scale instrument (1 = low, 10 = high) developed by the author. The purpose was to examine whether any of these variables had an effect on children's success." (Bilal, 2001, p. 124)

indicator

List of Indicators
Indicator # Indicator Label Indicator Statement
2 Energy efficiency "Energy efficiency of the library building"
4 Convenience of hours "Convenience of library hours to users"
... ... ...
(Mittermeyer, 1999, p. 16)

inductive reasoning

"As a method for developing categories, the researcher employed 'open coding'..., which is an inductive coding technique: '[Codes] emerge out of the data rather than being imposed on them prior to data collection and analysis'..." (Kim, 2000, 889)

"Coding categories were developed inductively, through iterative passes through the data as term pairs were sorted into content categories. There was no a priori scheme of categories." (Ross and Wolfram, 2000, p. 951)

informed consent

"Due to the school's Internet use Policy, children's parental consent to use the Internet was sought. Out of 90 invitations for participation, 30 consent forms were received. Of these, 25 children were willing to take part in the project." (Bilal, 2001, p. 122)

"In order to participate, students were required to give their consent by signing informed consent documents." (Jiao and Onwuegbuzie, 2001)

intensive interviewing

"The diary entries were used as a basis for in-depth exploration of participants' information behaviours in weekly interviews... Interview lengths varied from one-half hour to several hours, depending on the participant's interest and available time." (Julien and Michels, 2000, p. 5-6)

internal validity

"To ensure the internal validity of the findings of the qualitative interviews, formal member checking...was employed. At the beginning of the interview for the formal member checking, a short summary of the combined findings, which contained analytical categories and their definitions, respectively, was presented to each of the interviewees. After each interviewee was given a opportunity to look over the combined findings and comment on them, the interviewees were also asked to evaluate the accuracy of the findings that pertain to themselves." (Kim, 2000, p. 890)

interview

"Each advisor and doctoral student had a semi-structured interview for 45-120 minutes according to a pretested interview schedule. In addition to field notes, most interviews were tape recorded." (Covi, 2001, p. 1285)

"After the completion of the project, the 50 students who had submitted a completed project were interviewed using open-ended questions, and the interviews were audiotaped and subsequently transcribed. The questions were designed to gather data about the children's assessment both of the Web in general and their own searching experiences in particular." (Large and Beheshti, 2000, p. 1072)


L

law

"...Mooer's Law: An information retrieval system will tend not to be used whenever it is more painful and troublesome for a customer to have information than for him not to have it". (Austin, 2001, p. 607)

"Lotka's law...
f(k) = C
ka
In this expression k is a positive integer, denoting the number of publications, and f(k) is the number of scientists who have k publications included in the bibliography. The exponent a is a parameter to be determined from the data, and C is a normalizing constant..." (Kretchmer and Rousseau, 2001, p. 610)

longitudinal study

"...a series of five questionnaires over a six month period. Each questionnaire contained baseline questions about work and health during the preceding fortnight..." (Steward, 2001, p. 143)

"One particular securities analyst was identified as an information worker appropriate for longitudinal investigation of changes in an early career employee's perceptions of information seeking over time. ... Perceptions of the participant in the entry years of his career were compared with those 5 years later." (Kuhlthau, 1999, pp. 402-403)


M

mean

"...for each group the mean score for each of the 62 indicators (variables) was calculated." (Mittermeyer, 1999, p. 13)

"The mean number of terms in unique queries was 2.4." (Spink, Wolfram, Jansen, and Saracevic, 2001, p. 230)

measurement validity

"Users of IR systems frequently express satisfaction with poor search results and, perhaps, poor system performance. The evidence requires us to question the validity of these user assessment variables, for example, satisfaction with search results and perceived ease of use, as measures and predictors of actual search success and system performance." (Hildreth, 2001)

median

"The median was eight pages viewed per user. However, 28.6% of users examined only one page of results, i.e., because a page contains 10 ranked Web sites, about one in every four users look at 10 or less sites." (Spink, Wolfram, Jansen, and Saracevic, 2001, p. 229)

mode

Appendix C: Median, Mean, and Mode for Dead, Duplicate and Zero Links
Query\Serv Alta V Excite HotBot Infoseek Lycos
Dead median 1 0 1 3[2..5] 4.5[3..6]
Dead mean 1.53 1.07 1.80 2.87 4.27
Dead mode 0 0 1 2 5
... ... ... ...
(Leighton and Srivastava, 1999, p. 881)

multi-stage sampling

"Journals in mathematics, physics, psychology, and education were selected for this study... Because journal quality was a variable in this study, the top and bottom 20 journals in each field, as ranked by impact factor, were selected from each of the four fields..." (Harter and Park, 2000, p. 944)

N

nominal scale

Reasons for non-use of ICT facilities
Didnít know available No interest Donít know how Access elsewhere No help available Other
(Eve and Brophy, 2000)

"Subject analysis of the samples was based on BIC [Book Industry Communication] Standard Subject Categories, Level 2 (http://www.bic.org.uk/prodinf.html), grouped appropriately for public and academic libraries." (Chapman, Creaser, and Spiller, 2000)

nonresponse bias

"...since this survey was self-administered, it is likely that only those potential respondents who have had flow experiences were willing to participate in the study and report their experience. Further, since participating in the study required the ability to narrate personal experience, it is then possible that only those persons who were willing to expose their personal experience actually participated in the study." (Chen, Wigand, and Nilan, 2000)

null hypothesis

"Hypothesis H0 1: subjects with different educational and professional status do not differ in their mental models..." (tested using MANOVA, ANOVA, and Tukey tests) (Zhang and Chignall, 2001, p. 448)

O

operational definition

"The navigation path chosen by each subject for each task was compared across users. ... When at least six of the eight (75%) subjects selected the same path, the organization was judged to be the appropriate information structure." (Vaughan and Schwartz, 1999, p. 25)

"Printing, mailing, downloading, or saving ... are overt acts that the user takes when they perceive that the entry on the screen is worth keeping - for whatever reason. If, during a session, a user performs a search, displays a record from the search, and then takes one of these actions, it is an indicator that the user has found something relevant. We will take this to constitute a simple operational definition of whether the entire session was relevant." (Cooper and Chen, 2001, p. 814)

ordinal scale

"Frequency of use was measured on a six-point scale: 0 = no use; 1 = less often than or nearly once a month; 2 = two or three times a month; 3 = about once a week; 4 = two or three times a week; and 5 = almost every day." (Zhang, 2001, p. 634)

"The users' satisfaction with the search, i.e., the impression they got immediately after the on-line session was completed, but before they evaluated the printouts, was measured on a five-point scale: excellent, good[,] satisfactory, poor, bad." (Efthimiadis, 2000, p. 1000)


P

pie chart

http://libres.curtin.edu.au/libres11n1/Ocholla_files/image006.gif (Ikoja-Odongo, 2001, Fig. 2)

population

This study being a survey of Canadian cities, the first step consisted in defining the population and identifying eligible cities. According to the 1991 Census Data, 146 Canadian cities had a population of 25,000 or more... In the upper limit, only the city of Montreal and the former City of Toronto (before 1998) were excluded given the much larger size of their population.

The other criterion for the inclusion of a city was, evidently, that the city had a public library. But, more important for this part of the study was the requirement that: a) the library be administered by a board of trustees, and b) that the library had a formally organized Friends of the Library group.

(Mittermeyer, 1999, p. 6)

"Subjects were recruited from four populations: professional librarians and information specialists, graduate students, undergraduate students, and high school students." (Zhang and Chignall, 2001, p. 449)

posttest measure

"When questioned following the experiment, the subjects confirmed that the tasks were similar in style and complexity to ones they would normally experience." (Sutcliffe, Ennis, and Watkinson, 2000, p. 1213)

predictive validity

"This research was conducted in the Spring of 1998, and for some of the students it was their first contact with the Internet... Would the students' perception of the Web change as they became more acquainted with it, and if so, would they become more positive or more skeptical?" (Large and Beheshti, 2000, p. 1078)

pretest

"The process of developing the questionnaire included pretest interviews with 13 respondents. The questionnaire was revised after each interview." (Schamber, 2000, p. 735)

"A pilot focus group was conducted with a convenience sample of students from graduate classes at Indiana University. The pilot group allowed the investigators to practice the activities, check clarity of instructions, and check appropriateness of the techniques." (Vaughan and Schwartz, 1999, p. 591)

pretest measure

"In the tests where the user was allowed to use only the Web browser (pre-test), five tool-neutral questions were asked far each." (Heo and Hirtle, 2001, p. 669)

"The subjects' experience in the domain and information retrieval system was assessed by a pre-test questionnaire." (Sutcliffe, Ennis, and Watkinson, 2000, p. 1212)

primary sampling units

"Every month, six public libraries and six academic libraries each send in two random samples - ten titles at the cataloguing stage and ten titles at the ordering stage. The libraries are selected randomly for participation and each takes part for six months.... (Chapman, Creaser, and Spiller, 2000) (Libraries are the primary sampling units.)

purposive sampling

"A purposive sample for qualitative interviewing was selected from the population of faculty and graduate students at the Bloomington campus of Indiana University, who had published at least one scholarly e-article meeting the following criteria on the Web as of October 23, 1997:..." (Kim, 2000, p. 888)

Q

qualitative research

"This study employed both quantitative and qualitative inquiry methods... The qualitative method generated data from questionnaires and interviews." (Bilal, 2001, p. 122)

"This qualitative study, which explored the hyperlinking motivations of the authors with regard to individual hyperlinks used in their scholarly e-articles published on the Web, employed in-depth, open-ended, and semistructured interviews." (Kim, 2000, p. 889)

quantitive research

"This study employed both quantitative and qualitative inquiry methods. The quantitative methods provided empirical data about children's cognitive and physical behaviors in using Yahooligans! These behaviors were captured by using Lotus ScreenCam..." (Bilal, 2001, p. 122)

questionnaire

"A survey questionnaire was designed to collect data on scholars' demographic information, current frequency with which they used Internet tools and protocols, and means of obtaining access to various Internet tools and applications. Also requested were data regarding..." (Zhang, 2001, p. 633-634; for the complete questionnaire, see Zhang, 2001, p. 651-654)

"A participant profile questionnaire and a post-evaluation questionnaire gathered data to supplement the examination of the automatic QE [query expansion] results (Appendix A-1 and A-3)." (Greenberg, 2001, p. 408; for the questionnaires, see Greenberg, 2001, p. 413-414)


R

random assignment

"A total of 80 students at the University of Pittsburgh each having a minimum of 5 hr of Web experience were recruited as participants. Ten participants were randomly assigned to each experimental condition cell in a 4 X 2 between subjects design " (Heo and Hirtle, 2001, p. 668)

"The experiment...was divided into two parts, B1 and B2. In Part B1, the subjects were randomly slotted into one of four blends of community/personal filtering..." (Shepherd, Duffy, Watters, and Gugle, 2001, p. 156)

random sample

"An additional sample of 42 titles was selected randomly from the source journals in ISI's Arts and Humanities Citation Index (1993 print ed.), because the JCR does not include journals from the arts and humanities." (Harter and Park, 2000, p. 944)

ratio scale

"Speed is measured in terms of time, in seconds, that a user took to assess the relevance of a single document." (Tombros and Crestiani, 2000, p. 936)

"Time and success were combined in a measure of performance efficiency: the ratio of the number of successfully completed tasks to the time to complete these tasks." (Lazonder, Biemans, and Wopereis, 2000, p. 579)

reactive effect

"...questions about user confidence on the questionnaire showed that 28 users were confident in finding what they needed using the OPAC. Yet, during observations, this was manifested by only 14 participants. This raised the question of whether users gave responses that they thought were the ideal 'answers.' ... Finally, the researcher recognizes that the obtrusive nature of the observational and interview methods might have had some effect on the behavior of participants." (Slone, 2000, p. 765)

reliability

"...an inter-coder reliability check was used... for the coding of the data regarding classificatory decisions for bookmarks. A random selection of eight of the 15 surveys, representing 49 out of a potential 89 bookmark classification decisions, was coded by a second coder... Using Holsti's...formula for coefficient reliability, the selections of attributes made by the primary coder and the secondary coder corresponded 87.7% of the time..." (Gottlieb and Dilevko, 2001, p. 525-526)

"The agreement coefficient among the three independent coders was 83.6%, which is an acceptable level of intercoders' consistency, particularly for an exploratory study..." (Kim, 2000, p. 889)

replication

"The Web search tool study by Ding and Marchionini ...was selected as research to replicate because of its thorough documentation." (Nicholson, 2000, p. 725)

"The present investigation partially replicates, in the discipline of communication, the Shadish et al. measures with the intention of answering both the question they pose above (for what specific reasons was a particular highly-cited work cited?) and a corresponding question (do highly-cited authors differ from Shadish's sample of authors in the specific reasons given for citing their works?)." (Case and Higgins, 2000, p. 635)

representativeness

"The universities are not representative of every institution in their respective categories, but the sample as a whole reflects diverse working conditions within and across research universities." (Covi, 2000, p. 1286)

"Our response rate gives us some confidence in the representativeness of our sample. We compared respondents and nonrespondents in terms of whether they report an e-mail address in their contact information and in terms of the Carnegie Classification and National Research Council ranking of their institution... We find that...our sample...is skewed somewhat toward those who use e-mail and toward those at nonresearch institutions." (Walsh, Kucker, Maloney, and Gabbay, 2000, p. 1297)

response rate

"By...the cutoff date of the author survey, 179 researchers had responded to the survey request... With 201 potential respondents, the overall response rate was 89.1% (179/201)..." (Zhang, 2001, p. 634)

"...in a previous study this researcher found that preliminary contacts with the Mayor or his/her representative eliciting city council's participation, and a personal visit to the Mayor's office to deliver the questionnaires individually addressed to councillors resulted in an overall return rate of 77% for 35 municipalities visited [27 out of 35]..." (Mittermeyer, 1999, p. 7)


S

sampling

"Time periods for the study were chosen 2 weeks in advance of data collection based on each branch manager's suggestion of library 'busy' hours... All users who approached the on-line system during these times were invited to participate." (Slone, 2000, p. 758)

"Highly-cited authors were first identified, based on two recent studies of such authors in communication... The So and Funkhauser rankings were compared and two most highly-ranked authors common to both lists were chosen for the present study..." (Case and Higgins, 2000, p. 638)

sampling frame

"These fields were chosen because [among other things] sampling frames were available. ... By using membership directories, we get a representative sample of all types of institutions, those who use e-mail and those who do not, and those who do and do not do research." (Walsh, Kucker, Maloney, and Gabbay, 2000, p. 1296)

"The SSCI and AHCI are problematic as sampling frames, as they do not index citations to some journals in a discipline, and provide old or incomplete addresses for many of the citers. Even using association membership directories to confirm respondent addresses in the discipline of communication did not prevent 14% of the initial mailing from coming back as 'undeliverable'..." (Case and Higgins, 2000, p. 644)

secondary sampling units

"Every month, six public libraries and six academic libraries each send in two random samples - ten titles at the cataloguing stage and ten titles at the ordering stage. The libraries are selected randomly for participation and each takes part for six months...." (Chapman, Creaser, and Spiller, 2000) (Titles are the secondary sampling units.)

sequence effect

"It was our belief that if we had exposed users to condition C first, they would have been negatively biased because of the complexity of that condition. Therefore one can argue that the results for condition C reported in this paper reflect an optimistic or an 'upper bound' view of user performance." (Tombros and Crestiani, 2000, p. 935)

sequential sampling

"During February 1997, the test suite developer recorded the verbal request of every reference question that he was asked in which the patron specifically requested that the Internet be used as a source of information. We stopped after 10 queries were obtained." (Leighton and Srivastava, 1999, p. 872-873)

snowball sampling

"Participation was obtained... using a handshake sampling method, since a high degree of trust was needed in order for participants to submit willingly to the in-depth, time-consuming, and personal exploration of their information behaviour asked of them." (Julien and Michels, 2000, p. 4)

"A snowball-sampling method was considered to be an appropriate method of participant selection given the sensitive nature of the topic... As community members found out about the study, and if they wanted to take part, they approached the researcher either directly or through an intermediary." (Sligo and Jameson, 2000, p. 862)

standard deviation

"The time lag between the publication date of the original item and the citing item ranged from 1 to 30 years, with a mean of 12.3 years, a median of 12, and a standard deviation of 6.1 years." (Case and Higgins, 2000, p. 642)

"The subjects' level of domain expertise was indicated by their grade in Dutch literature. On a 10-point scale, the overall mean for this measure was 6.0... The WWW-experts' mean score for domain expertise was 6.4 (SD = 0.9). The domain expertise of WWW-novices was lower (M = 5.8, SD = 0.9)..." (Lazonder, Biemans, and Wopereis, 2000, p. 578)

stratified sampling

"The sample was stratified to include queries with advanced search features." (Spink, Wolfram, Jansen, and Saracevic, 2001, p. 233)

"...we used the World of Learning and university websites to create a database list of all business school faculty employed at those intstitutions. ... A stratified (stratified by academic institution) random sampling was then drawn from our database.... (Speier, Palmer, Wren, and Hahn, 1999, p. 539)

survey

"...requests were sent to the editors of the eight journals ... Specifically, the requests asked whether there were or there had been in the past 10 years any editorial policies regarding how e-sources should be referenced. Furthermore, the following information about each in-press article was requested: ..." (Zhang, p. 633)

"The primary data come from a mail survey of scientists in experimental biology, mathematics, physics, and sociology." (Walsh, Kucker, Maloney, and Gabbay, 2000, p. 1296)

systematic sampling

"Beginning with the first article carried on the University's news server, we extracted every tenth article from two related Newsgroups, to a total of twenty articles per group." (Brown-Syed and Morrissey, 1999)

T

t test

"The Student's t-test was...used to test the difference in means between 1996 and 1999." (Davis and Cohen, 2001, p. 311)

"A significant difference among the four conditions lies in the average time taken to assess the relevance of one document, in particular between conditions V and C (significant at the 2% level for a two-tailed T- test), and T and C (significant at the 5% level for a two-tailed T- test)." (Tombros and Crestiani, 2000, p. 938)

test statistic

"Comparing those items which received 11 or more cites with those that received ten or fewer cites showed significant statistical differences (t = 2.773; df = 54, p = 0.008) on only one item: 'this reference reviews prior work in this area.'" (Case and Higgins, 2000, p. 642) (t is the test statistic)

"Experts also produced significantly higher performance success scores (t(24) = 2.52, p < 0.01...), indicating that they successfully completed more tasks than novices." (Lazonder, Biemans, and Wopereis, 2000, p. 579)

theory

Dervin's sense-making theory: "information is not external to humans"; "information is subjective"; "information seeking and use are activities that people undertake to construct...sense"; "sense-making behaviour is partly situationally dependent and predictable". (Julien and Michels, 2000, p. 3-4)

"Four concepts constitute the theory of normative behavior: (1) social norms; (2) worldview; (3) social types; (4) information behavior." (Burnett, Besant, and Chatman, 2001, p. 537)

triangulation

"Complementing the citation data, this study also collected data about how e-sources have been used during the research process." (Zhang, 2001, p. 644)

"The generalizability of the findings of the qualitative interviews would be improved by using 'between-method triangulation'... In the second phase of the larger study, which employed qualitative and quantitative methods in combination, a quantitative mail questionnaire survey was used to investigate the hyperlinking motivations of scholarly electronic journal authors in a wide range of disciplines." (Kim, 2000, p. 890-891)

type-I error

"...a series of independent t-tests, using the Bonferroni adjustment for Type 1 error, revealed that the study habits exhibited by the graduate students were statistically significantly higher than that exhibited by the undergraduates in all five studies...." (Jiao and Omwuogbuzie, 2001)

U

unit of analysis

"These data on information sources are based on 79 searches for help or information, since in nine of the 88 transcripts [of interviews with 9 participants] participants' responses strayed significantly from the interview questions posed. ... In 45 of the 79 interviews (58%), participants turned first to personal sources of help with their questions." (Julien and Michels, 2000, p. 7-8) (The unit of analysis here is interviews or searches, not participants.)

voluntary participation

"After the searches were completed, the participant was required to complete a post-search questionnaire.... Only after completing the questionnaire would the participants receive their ten dollar payment. This may explain why there were no missing values on these instruments." (Hildreth, 2001)

References


Home

Last updated October 1, 2001.
This page maintained by Prof. Tim Craven
E-mail (text/plain only): craven@uwo.ca
Faculty of Information and Media Studies
University of Western Ontario,
London, Ontario
Canada, N6A 5B7