The Role of the Principal in an Information Literate School Community: Design and Administration of an International Research Project
Dianne Oberg, Professor, Department of Elementary Education, University of Alberta, Canada; Lyn Hay, Lecturer, School of Information Studies, Charles Sturt University, Australia; James Henri, Senior Lecturer, School of Information Studies, Charles Sturt University, Australia
Support of the principal is a key factor in the implementation of effective programs in schools. An international study of the principal’s role in developing and supporting information literate school communities was conducted in Australia, Canada, Finland, France, Japan, Scotland, and South Korea. The study sought to inform the efforts of principals and teacher-librarians throughout the world seeking to develop information literate school communities. Such school communities place a high priority on the mastery of information-use processes by both teachers and students ; that emphasis on information literacy is reflected in policy, benchmarking, funding, and evaluation. One of the unique features of this international study was the use of online data colleciton and analysis techniques. In this paper, the researchers describe the design and administration of the study and explore the methodological issues involved. This information will be of use to researchers interested in replicating the study or in designing a similar study.
Development of the International Study | Principal Support in the Literature | Research Method | Online Approach to Data Collection | Data Entry and Analysis | Reporting the Findings | Acknowledgment | Notes | Works Cited
During the 1990s, the authors of this paper have been developing a program of research investigating the role of the principa (
1) in relation to school library programs and services. In 1996, they obtained funding from the International Association of School Librarianship (IASL) (
2) and the International Federation of Library Associations and Institutions (IFLA) (
3) for an international study of the role of the principal in developing an information literate school community. The study has been carried out in seven countries: Australia, Canada, Scotland, Finland, France, Japan, and South Korea. Up to this point, research findings have been disseminated primarily through conferences of IASL and IFLA. Because of the international context of this research, the authors have chosen to use the terminology most frequently used in the participant countries and in the research instruments used for the study, that is, “teacher-librarian” and “school library.”
This program of research had its roots in qualitative work done by Dianne Oberg and Linda LaRocque in Canada (LaRocque and Oberg 1991; Oberg 1996). Lyn Hay and James Henri then carried out a study in Australia, based on the Canadian study. Findings from the Canadian study were presented at the 1990 conference of IASL (LaRocque and Oberg, 1990), and findings from the Australian study were presented at the 1995 IFLA Schools Section Open Session and in Australia (Hay and Henri 1995; Henri and Hay 1996). The qualitative studies, conducted by Oberg and LaRocque and by Hay and Henri, provided analyses of the ways in which principals working within information literate school communities are able to support the work of teacher-librarians. The projects also identified the methods used by teacher-librarians to involve the principals in the development of effective school library and information services. The Canadian project involved five schools in Alberta while the Australian project was undertaken in six schools in New South Wales.
While the qualitative studies had provided in-depth understanding of a small sample of schools in two countries, the researchers felt that it was important to test the validity of these findings through a quantitative study. Having identified the factors of influence and support that exist between the principal and the teacher-librarian though qualitative studies, the researchers undertook the development of quantitative instruments to test the existence of these factors across a broader range of schools and in a larger number of countries. Thus, the international research project titled “The Role of the Principal in an Information Literate School Community” was developed and implemented. This paper focuses on the development and administration of the international research project.
The international study involves a quantitative investigation, surveying both principals and teacher-librarians about principal support, making use of data from the original qualitative studies. Involvement of other countries in the research began at the 1995 IFLA conference, where Hay presented a paper on the six-school study in Australia. Based on expressions of interest at that conference, funding was sought from both IFLA and IASL. For the 1997 IFLA conference, Hay, Henri, and Oberg organized a full-day workshop. Four papers were given on the research related to the role of the principal in an information literate school community (Dogg Hafsteinsdottir 1997; Henri and Hay 1997; Moore 1997; Oberg 1997), and a workshop was held for members of the International Research Reference Group (IRRG) representing the seven countries involved in this international study (see
The role of this group was to: (a) provide input and advice regarding the adaptation and translation of the quantitative and qualitative instruments for their country; and (b) plan and administer the procedures for data collection, analysis, and reporting of findings for their country.
[ Back to Top]
Over the past three decades, there have been frequent references in the professional literature of teacher-librarianship to the concept of principal support, but there have been fewer references in the research literature. Shields (1987) and Charter (1982) found that principal support was critical to the development of school library programs. Dekker (1989) found that school district administrators were also important in enabling principals to support library programs in their schools. Corr (1979) and Turner (1980) found that principal attitude was positively correlated to program implementation. Hellene (1974) and Yetter (1994) found that principal support for the school library program involves such things as encouraging its use by teachers and students, integrating the program into the curricular work of the school, and providing flexible scheduling.( 4) Wilson, Blake, and Lyders (1993) found that many principals were hampered in their support for school libraries by lack of knowledge about the management and function of school libraries.( 5) Table 2 (Oberg 1996) provides a summary of the professional and research related to the concept of principal support for the school library program, a key factor in the development of an information literate school community.
Design of the Instruments
Questionnaires, based on the interviewee data fields used and the key factors resulting from the original qualitative studies, were developed and tested in Australia. The piloting of these instruments was conducted in Australia using standard hard-copy questionnaires. The Internet was used to distribute the pilot instruments to volunteer members of the Australian discussion group OZTL_NET; however, the inability of many respondents to translate e-mail attachments resulted in the faxing and/or snail-mailing of instruments to the majority of the participants in the piloting.
Two model questionnaire sets—one for principals and one for teacher-librarians—were developed. The three instruments in each of the questionnaire sets included both closed-choice and open-ended questions. The closed-choice questions in Instrument 2 employed a five-point scale, with a zero weighting for the additional category “cannot comment.” The traditional five-point scale was rejected because the instruments were lengthy and it was felt that there might be an interest in over-using a mid point.
One goal of the researchers was to review their overall research design and methodology in light of their experience with this international project. Discussed below is the design of each of the three instruments used in the study and some of the problems encountered in the data collection process. Also described is the online approach to data collection that was used in the study and the advantages and challenges of such an approach is explored.
Instrument 1: Demographics
Instrument 1 was designed to identify demographic variables for each of the country samples, including the personal and professional characteristics of the principals and teacher-librarians and the characteristics of individual schools. Principal and teacher-librarian respondents were required to complete different versions of Instrument 1. Both principals and teacher-librarians were required to provide their own personal and professional details. In addition, the principals were asked to provide some whole-school data, while the teacher-librarians were asked to provide specific school library resource center data. The researchers decided to split the demographic data across both versions of Instrument 1 to avoid duplication of school-based demographic data and to reduce the data input burden for both principals and teacher-librarians. Examples of the Australian online versions of Instrument 1 can be found at
Table 3 summarizes the demographic variables for the principal version of Instrument 1. Question 1 contained closed-choice questions in which respondents were asked to select a category or value that best defined their individual school type. A standardized set of values—“government,” “community” and “private"—was devised to define the different types of school systems. The labels of these values were then edited into the preferred terminology of the educational system for each country. For example, the Australian principal instrument used only the two values of “government” and “non-government,” whereas the Finland instrument used all three standardized values (see table 1). The Finland version also asked for two additional types of values: (a) whether the school was identified as a “specialty” school (i.e., arts, sports, etc.), and (b) whether the school was included in the experimental school reform movement—both specific to current educational administration in Finland.
A set of values for the grade variable type was also customized to reflect the climate of each country’s school system, and included an additional open value of “Other” to ensure that all grade ranges were identified correctly. For example, the Canadian instrument listed the range of grade values as “K–6,” “7–12” and “Other,” whereas the Japan instrument used the values “1–6 elementary,” “1–3 middle school,” “1–3 high school” and “Other.” The customization of Principal Instrument 1 for each country lead to a total of eight grade level values across the seven nations: K–6, 1–6, K–10, 7–9, 7–10, 7–12, 10–12, and 11–12. Rather than collapse these values into a smaller number of standardized ones, researchers kept the integrity of each for the purpose of permitting more meaningful future analyses for individual countries and for cross-country comparisons.
Questions 5 and 6 were also closed-choice questions for the variables of principal age and gender. Respondents were asked to select from the age-range scale of 20–29, 30–39, 40–49, 50–59 and 60+ years, and from the gender scale of female or male. All other questions were answered using open fields. Some countries encountered problems with the open-field design of the second variable in question 2, where principals were asked to identify the number of days or hours per week worked by part-time teacher-librarians (i.e., those teaching fewer than five full school days per week). Some respondents indicated a number without indicating whether that value represented either a unit of one day or a unit of one hour. The instrument designers had not foreseen the number of ways in which these data could be presented. For example, a teacher-librarian who teaches three days each week could be reported as either: (a) 3 days; (b) 0.6 of a position, or (c) 18 hours per week. This was the major flaw in the principal version of instrument 1. While the majority of data could be coded correctly, some data were ambiguous and could not be used as an accurate measure. For instance, the problem did not arise for South Korea because only those schools employing a full-time teacher-librarian were surveyed; however, the data for this variable were invalid for Japan.
Questions 3 and 4 required principals to record the total number of Internet connections in the school and the total number of Internet access points (or computer terminals) in the school. These questions were designed to identify the extent of Internet access throughout the school. However, both questions caused some confusion for respondents in that some respondents could not make a distinction between the terms “connection” (i.e., the number of connections via a modem pool or dial-up connections available at any one time) and “access points” (i.e., the number of terminals throughout the school that allow users to access the Internet). As a result, the data from questions 3 and 4 could not be reliably used. However, the term “access points” seemed to be better understood by respondents in most countries and, therefore, the question 4 data seem to reflect more accurately the level of Internet access for each country.
The remaining six questions on Principal Instrument 1 focused on principals’ professional education and experience. For principals’ academic qualifications (question 7), the assigned value range consisted of “Bachelors Degree or undergraduate equivalent,” “BA/ Diploma Ed/ Honours,” “Masters or postgraduate degree,” and “PhD.” Questions 8, 9, and 10 were designed to determine how many years the principal respondent had been teaching before taking on the position of principal, how many years they had held their current school position as principal, and the total number of years that the respondent had functioned at the school executive level. Question 11 sought to identify the number of teacher-librarians the respondent had worked with since becoming a principal, while Question 12 required respondents to list the names of professional associations to which they belonged.
The teacher-librarian version of Instrument 1 was designed to identify demographic variables of the school library resource center (SLRC) and the teacher-librarian position, as well as characteristics of the individual person holding the position of teacher-librarian in the school. Table 4 summarizes the demographic variables for the teacher-librarian version of Instrument 1.
Questions 1 and 2 were designed to identify the level of Internet access in the school library resource center compared with that in the overall school (based on questions 3 and 4 in Principal Instrument 1). Once again there was some confusion among respondents about the difference between “connections” and “access points.” Questions 3 and 4 were closed-choice questions for the variables of teacher-librarian age and gender. Respondents were asked to select from the same age and gender scales used for principals. The remaining nine questions on this version of Instrument 1 focused on the teacher-librarian’s professional education and experience. An additional value of “Certificate/TAFE/Trade” was included in the teacher-librarians’ academic qualification (question 5) range to that of the principal range. Questions 6 and 7 asked how many years the respondent had been teaching before taking on the position of teacher-librarian and how many years they had held their current position as teacher-librarian. Questions 9 and 10 were designed to determine the activity of the teacher-librarian in informal leadership ("pseudo-executive") positions and formal executive positions. Question 8 was designed to identify whether the teacher-librarian position had been filled using a merit selection process or allocated according to internal transfer coordinated at the educational system level. Questions 11, 12 and 13 were used to identify the teacher-librarian’s level of professional-development activity based on membership in professional associations and discussion lists and reading of professional journals.
Instrument 2: Perceptions and Beliefs
Instrument 2 was designed to identify the level of principal support for the school library program and the teacher-librarian. Principals answered 50 questions and teacher-librarians answered 53 questions using five-point rating scales. Instrument 2 was divided into two parts: (a) perception factors, and (b) belief factors. Examples of the Australian online versions of Instrument 2 can be found at
Part A: Perception Factors
Principals and teacher-librarians completed identical versions of the 31 perception questions in Instrument 2. First, respondents were asked to rate the level of attention they perceived the principal to give each item at present using the rating scale, “A Lot–Some–Little–None, or Cannot Comment.” Respondents were then asked to rate (using the same scale) the level of attention they would like to see the principal give each item in the future. Table 5 groups the perception items of principal support surveyed in part A according to the four types of principal support identified earlier in table 2.
Part B: Belief Factors
Principals and teacher-librarians completed different versions of belief statements in Instrument 2. While both respondent groups were asked to indicate the strength of their belief for each of the items using the scale “Strongly Agree–Agree – Disagree–Strongly Disagree–Cannot Comment,” the principal instrument consisted of 18 belief statements whereas the teacher-librarian instrument consisted of 21 belief statements.
Table 6 lists the belief statements (in a shortened version) regarding principal support surveyed in part B. These belief statements were designed to indicate the strength of and alignment between principal and teacher-librarian beliefs about the roles of principals and teacher-librarians in developing and supporting an information literate school community. The belief statement responses can also be used to shed more light on the results found regarding principal attention in part A. In retrospect, asking the additional questions of both principals and teacher-librarians would have increased the value of the data generated from these questions. For example, in relation to the teacher-librarian role being good preparation for the principal position, the principals might have had quite different views from those that were expressed by the teacher-librarians and this is information that would enrich teacher-librarians’ understanding of how their role is perceived by the principal.
Instrument 3: Additional Information
Instrument 3, composed of fifteen open-ended questions, was designed to elicit additional information from principals and teacher-librarians in a qualitative format. Nine of the open-ended questions were the same for principals and teacher-librarians. These questions invited respondents to make comments related to the strengths and challenges of the school library, the contributions of teacher-librarians to teaching and learning, the nature of information literacy, barriers to integration of information skills, the promotion of the school library, and the respondents’ roles in developing and supporting an information literate school community. Teacher-librarians were asked two extra questions related to ways in which they maintain their credibility as teacher-librarians and ways in which their principals could provide them with additional support. Examples of the Australian online versions of Instrument 3 can be found at
In retrospect, asking principals the two extra questions would have provided useful information about teacher-librarian credibility and about the nature of principal support. The level of response to these open-ended questions varied from country to country, with the Canadian and Australian respondents providing the most lengthy and detailed comments. This raises some interesting questions. The answers may be related to the fact that the open-ended questions were initially framed by the Australian researchers. Thus the questions may have been expressed in terms that were more meaningful for the Australians and Canadians, two respondent groups with educational systems more similar than those of the other countries.
[ Back to Top]
In-school research is typically slow. Gaining approval to conduct the research involves meeting the demands of approval protocols: completing numerous forms, providing sample instruments, indicating timelines, and dealing with many levels of authority. In addition, the candidate respondents, especially principals, are very busy professionals. School and district authorities are often reluctant to approve proposals that will add demands to already overworked professionals. It was with these factors in mind that the decision was made to transfer the administration of the questionnaires from a snail-mail, paper-based approach to an online, Web-based approach.
Lyn Hay was the IRRG Web site coordinator and consulted with IRRG members to adapt and translate the questionnaires to interactive HTML forms. Each country was given its own home page on the IRRG Principals Project Web site where the online version of the questionnaire was posted. E-mail addresses for IRRG members and the Web site coordinator and Internet Special Project Group that designed the data collection tools were provided for those respondents with questions. Figure 1 illustrates the home page of the IRRG data collection site. Figure 2 illustrates the Australian home page. The Australian format became the template for all other country home pages.
Each school was assigned a School Identification Number (SIN), which was an essential requirement for online data entry. The same number was assigned to both the principal and the teacher-librarian of a school, ensuring that the data sets from each could be electronically matched and manipulated in preparation for the data analysis phase. Three instruments were used to collect data. Instrument 1 included a combination of pull-down menu selection of set fields, and short and open-ended question fields. Instrument 2 collected data using a series of pull-down menus. Instrument 3 collected responses to twelve open-ended questions that were entered using a series of open window fields. All raw data was tagged by the SIN and the instrument number. Once the respondent completed and submitted the questionnaire, a special page confirmed its successful receipt.
The online questionnaires employed a simple Common Gateway Interface (CGI) script to capture data in a form that could be processed by Standard Generalized Markup Language tools (see
www.oasis-open.org/cover/sgml-xml.html). The CGI scripts were written in Python (see
www.python.org), a programming language simpler than the popular scripting language, Perl (
www.perl.org). Python was also used to convert the questionnaire data into a suitable form for processing.
Evaluation of the Online Approach
The main challenge in designing the online data collection site was assigning unique field names to each question in the complex questionnaires. In future studies, it would be preferable to write a script to generate the HTML questionnaires. A questionnaire could be marked up according to an Extensible Markup Language (see
www.oasis-open.org/cover/xml.html), DTD, and a script written to generate the final online document. XML is a simplified version of SGML designed for online applications. It is likely to replace HTML as it is much more versatile.
Just as the administering of a traditional snail-mail based questionnaire is fraught with possible delays (e.g., postal strikes, nondelivery, and incorrect delivery), so too is an online product. This case was no exception and there were a number of minor problems typically associated with the online operation. The server was subject to several power failures (as part of ongoing building construction at CSU) and outages (as networks systems were being improved). The unique SIN would have been more reliable had it included a check digit that would have guarded against a respondent entering an incorrect number. Although the snail-mail versions of the instruments were tested through the pilot process, the online versions received only in-house testing. Previous experience with programming of online instruments had ensured that the system was robust enough to cope with the typical errors (although the original error message was the somewhat facile “programmer error"). Error reports from some respondents indicated that more robust public testing of the online versions would have been beneficial.
One potential problem, associated with the online version of Instrument 2, was related to the choice of default. The default in the online version was set at “A Lot” and “Strongly Agree” with the thought that this would force respondents to make a choice. However, overlooked was the result that a respondent who ignored a question would (by default) still enter a value. Fortunately, a careful inspection of the data showed that this did not happen. However, in the future, the default will be set at “Please Select.” Another way to address this problem might be to replace the pull-down menus with radio buttons, which would also save respondent time and which would be more user-friendly to Internet novices.
Hard returns entered by respondents when completing open-ended responses also caused problems for data importation in preparation for data analysis. A hard return was read by the program as being a discrete, new piece of data when in actual fact it could have been one point among several entered as an answer for a single question. On identification of this problem, a script was written to correct this problem.
In addition to the problems associated with lack of testing, some additional problems with the online questionnaires should be mentioned. Many respondents did not have sufficient online experience to independently enter all of the required data and submit their surveys successfully. The online approach is dependent upon the robust nature of an individual school’s Internet connections. If a connection failed before all data were entered on an instrument, those data were lost. Should this happen repeatedly to a respondent, it would be unlikely that she or he would continue attempting to submit the data. If a respondent did not enter a SIN, the completed instrument could not be submitted. (While this represents a frustration, a submitted instrument without a SIN would prevent pair matching and would therefore be less valuable.)
A tabular summary of all data entered was monitored via a “Principals’ Survey Submission List” Web page.
Figure 3 illustrates the effectiveness of the data collection script in monitoring and managing the online data collection process regarding pair matching.
This allowed individual country coordinators and the project managers to monitor the percentage of data collected and allowed simple identification of missing instruments and SINs that had not been submitted (e.g., both principal and teacher-librarian respondents at the schools with SINs ca0005, ca0020 and ca0035 have submitted all three instruments, whereas the teacher-librarians at schools ca0006, ca0032 and ca0037 have submitted all three instruments but the principals at these schools have failed to submit any instruments. This facilitated the chasing of outstanding surveys from schools. In the Canadian study, in a province where Freedom of Information and Privacy legislation limits researchers’ matching of respondent names with their responses, the Principals’ Survey Submission List, was included in the second reminder letter to schools so that schools could check their own submission status. The success of the project hinged, to some extent, on the successful completion of instruments by the both teacher-librarian and principal of each participating school. The involvement of both allows for analysis on paired responses. If only one of the pair returned their data other analysis could be successfully undertaken, but the value of the study would be diminished somewhat.
Notwithstanding the problems noted above, the use of a Web-based approach to data collection has a number of significant advantages and enormous potential for future large international collaborative research projects. Perhaps most importantly, the approach allows the standardization of survey instruments and coding of data across countries. Likewise, all data from the participating countries can be collected on one server.
In addition, data collection methods can be standardized across participating counties, which allows data to be imported into the SPSS data analysis program. This is particularly useful because it facilitates standardized data testing and analysis across all countries and allows for simplicity in future comparative data analysis. These commonalities enhance the management of the project and enable ready monitoring of the progress and the timely identification of problems affecting all participating countries or arising from individual countries (if any).
The time required for data collection is reduced because respondents enter data directly to the server rather than onto paper that requires a third party to rekey the data. When funding is an issue, this is of special benefit. In addition, all data entered via the Web are automatically formatted, ready for importing into a data analysis program.
Respondents were able to submit each of the three survey instruments separately, and even to submit parts of an instrument over time. They thus had the flexibility to complete the survey in several sessions rather than sitting at their PC for a substantial block of time. This approach was evident in a number of cases where subjects submitted half an instrument and then came back later and submitted the remaining data. Using the SIN and instrument tag, data collectors were able to successfully match the two sets of data. This flexibility in data input may have enhanced the qualitative data entered in the third instrument. This also may help to explain why the Canadian respondents, the majority of whom responded online, provided the most lengthy and detailed responses to Instrument 3. An example of compiled data as a text file for Principal Instrument 3 can be found at http://farrer.riv.csu.edu.au/principal/PR3_au.txt
The online approach provides advantages to both the respondents and the researchers, particularly in a multicountry project where a number of languages were employed. The significant reduction in data entry time and the collection of all data together at a convenient point that enables a range of time savings and enhancements cannot be over-valued. In this study, however, this cutting-edge approach did create challenges for those respondents who were not regular online users. Improvements in off-the-shelf software will enable enhancements to the approach. These improvements would allow researchers to generate an online questionnaire by typing the questionnaire into a word processor using special markers; the data entered into the questionnaire by the respondents could be automatically prepared for processing using SPSS.
[ Back to Top]
Each IRRG member was responsible for the collection of data in their country and for the entry of those data via the Web database at the School of Information Studies, Charles Sturt University (CSU). Respondents had been asked to submit their responses via an online version of the questionnaire if possible. There was variation in data collection and data entry across the seven countries, and in no country was data collected and entered by electronic means alone.
In five of the countries, the data were collected using paper questionnaires. In four of these, the researchers then entered the data onto the CSU Web site. In non-English speaking countries, this second step also involved translating non-numeric data into English. In one non-English speaking country, this additional step proved too daunting; the data was never entered onto the CSU Web site, and was instead analyzed by a local research agency. In Canada and Australia, some respondents entered their data directly onto the CSU Web site while others completed paper questionnaires that were then entered onto the country’s Web site by the researchers. The Canadian respondents had the highest rate for data entered directly onto the CSU Web site (95%; 94 out of 99 respondents).
The quantitative data from the CSU Web site were analyzed using SPSS by Hay and Henri at CSU. Frequency analysis was used to get an overall picture of the data, and t-tests were used to check for significant differences between the responses of principals and teacher-librarians. The qualitative data from the open-ended questions were analyzed using a framework and procedures developed by Oberg at the University of Alberta.
Data Analysis for Instruments 1 and 2
The data analysis process for the quantitative Instruments 1 and 2 was developed using the Australian data. As each country’s data were analyzed, a comparative coding masterfile was developed to accommodate the additions or changes made to instruments of individual countries. A coding sheet was developed for coding of the demographic data from Instrument 1 for each country. Variables were created in SPSS for Windows 3.1 beginning with SIN, TL_PR (Teacher-librarian-1 or principal-2), and country. Variables for perceptions, both future and present, in Instrument 2 were created for each question (i.e., q1f, q1p, q2f, q2p) and belief questions (i.e., q32b, q33b). The number 999 was entered for missing values section under each variable. Each of these variables were then labeled under the label sections, so that coding is explained under each variable by doubleclicking on the variable label in the grid or data file of the SPSS.sav file. All other settings remained as the default settings. All perception questions in Instrument 2, both present and future, were coded in the following manner: 0 = no comment; 1 = none; 2 = a little; 3 = some; and 4 = a lot. Beliefs were coded in the same manner: 0 = no comment; 1 = strongly disagree; 2 = disagree; 3 = agree; and 4 = strongly agree.
Frequency analysis was carried out on Instrument 1 data to provide useful data for reporting, including quick and tangible figures, and a summary of information of nominal and ordinal scales. An initial frequency analysis was carried out selecting minimum and maximum values to be shown in order to check that all data were entered correctly and that 999 had been coded as a missing value and would not be included as a valid number, and therefore used in analyses. Any errors to data entry were corrected at this stage. The data file was then split so that all analyses would be run for teacher-librarians and principals as two separate groups. The File Split function was used under the variable TL_PR.
Valid percentages were used in giving the frequency data. Thus all 999 responses were excluded from the calculations, e.g., if there were 68 respondents and of those ten were missing (999 codes), the percentages were calculated from the 58 respondents who had valid responses for that variable. Percentages were then recorded for the report in order of Teacher-librarian variables, Principal variables and then School variables. Percentages of each response for each variable were recorded (e.g., for gender, it might be that 59% of teacher-librarians were female and 41% were male).
Descriptive data (mean and standard deviation) were used to provide a quick summary of the present, future, and beliefs data (ordinal data, e.g., 2 is more than 1 in value). Due to the coding and the later use of present, future, and belief data in analyses, it was more appropriate to use mean (average response) rather than share (percentage) for each code (e.g., how many answered “1 - none” to question 1 future, “2 - a little” to question 1 future, and so on). The data were still split at this stage so that teacher-librarian and principal data were analyzed separately. Mean and standard deviation responses were tabulated for each present and future question in one table. Teacher-librarian data were presented first. Average responses to each question were then presented in written form for present and future. Mean and standard deviation responses for principals were then tabulated in the same format as that for the teacher-librarians. Instead of a complete rundown for principals, a short paragraph summary was given for the teacher-librarians responses.
Belief responses were tabulated and presented for the teacher-librarian data followed by a summary of the beliefs that the teacher-librarians as a group believed were accurate and then those which they believed were inaccurate or less than accurate. Any mean greater than 3.0 was seen as being in clear agreement with the belief. Any mean less than 2.5 was seen as being in clear disagreement with the belief. Belief responses were tabulated and presented for the principals followed by a summary of responses that differed from those given by the teacher-librarians.
T-tests were carried out to compare present and future perceptions. A general standard of p<.001 was set for significance, because of the large number of tests tabulated. Setting a low significance level helped prevent a chance difference being mistaken for a significant difference. Dependent t-tests were run for the present versus future questions, first for teacher-librarians and then for principals. The tests compared responses to two different questions, rather than two different individuals’ responses to the same question. Results were tabulated giving the mean, standard deviation for present and future individually and the p value for each question that was significantly different from present time to future time spent. A written summary was then given of the questions that were significant (e.g., “Teacher-librarians believe their principals should be spending more time seeking feedback from staff about their impressions of the quality of the LRC services than they currently do.").
Independent t-tests were carried out to compare the two different independent variables of TL and PR for each question (e.g., q1f). Independent t-tests were carried out on all present, future, and belief questions, and significant results were presented. Again significant results were set at p</= .001. Results were then tabulated and a written summary given of the individual questions on which the principal and teacher-librarian differed. Tabulated results included the Levenes f value, Levenes p value, t-value, degrees of freedom, p value, TL mean and standard deviation, and PR mean and standard deviation. Independent t-tests were also carried out on the following variables: number of years in current position, age, gender, number of years in executive positions, number of years teaching prior to current appointment, qualifications, and number of professional association memberships. Significant results of these comparisons between the principals and teacher-librarians were tabulated for the present, future, belief questions, and a summary was provided.
Variables called “Present,” “Future,” and “Beliefs” were created by totalling respondents’ answers to the 31 questions in each of the respective categories. Totals were recorded for each variable, so the result would be out of 124 (31 x 4). Independent t-tests were carried out for these three variables comparing the principal and teacher-librarian for each one to see if they significantly differed from each other. All results were tabulated and a written summary provided.
Data Analysis for Instrument 3
Responses to the open-ended questions on Instrument 3 were analyzed using a content analysis approach. Content analysis involves a lengthy process of reading and re-reading responses, noting their content, identifying themes or categories according to the content, and then grouping and re-grouping the responses within the themes or categories. This interpretive process began with the reading of all responses in order to get an overall sense of the data. Each open-ended questions was then analyzed. For example, principal responses to question 3 were read and the the ideas within the response were written down. Themes were identified and the ideas were grouped under the themes. From this, the frequency of ideas could be seen and the dominant themes could be identified. This process was then repeated for the responses of the teacher-librarians to the same question. The NUDIST*QSR software program, designed for use with textual qualitative data, was used to gather together responses to each open-ended question by country. Responses were gathered separately for principals and teacher-librarians. The software was also used to gather pairs data, that is, the responses of the teacher-librarian and the principal for each school where both principal and teacher-librarian submitted responses to the questionnaires.
At this point, the responses to the open-ended questions by teacher-librarians and by principals in each country have been analyzed. In future, it is anticipated that researchers in some of the countries will want to analyze the responses to Instrument 3 provided by teacher-librarians and principals in the same school (pairs data) to identify possible patterns related to alignment of beliefs, perceptions, and goals for an information literate school.
[ Back to Top]
Reporting the Findings
The project has generated an immense amount of data, and in-depth analysis will take considerable time. By the end of July 1998 the first phase of the statistical analysis of the quantitative data was completed and the data analysis reports were shared with researchers from the participating countries. At that time, as well, the researchers received reports of the qualitative data organized by question number. Each member of the IRRG was responsible for compiling a preliminary report, based on findings from both qualitative and qualitative data analysis, for their own country. The presentation of these reports comprised the agenda for the morning workshop held at IFLA ’98 in Amsterdam, The Netherlands—Session 158B for the IFLA Section of School Libraries and Resource Centres on "The Role of the Principal in an Information Literate School Community: An International Research Panel." The papers from this session are available at http://farrer.riv.csu.edu.au/principal/survey/report.html.
Further analysis of the qualitative data was required to explore the forms of support for teacher-librarians offered by principals; the types of actions taken by teacher-librarians to develop principal support; the strategies implemented by principals and teacher-librarians in developing information literate school communities; and the professional development needs of principals and teacher-librarians with respect to developing an information literate school community. Here also the in-depth analysis of the pairs data, that is, the responses of principals and teacher- librarians working together in the same schools, may provide valuable insights into the ways in which they work together and the factors that support and limit their collaborative work. The contribution that this international study makes to the development of information literate school communities will be seen where data from the individual country studies are analyzed to reveal the patterns of collaborative work between principals and teacher-librarians within each country’s unique educational context and culture.
Analysis of the quantitative data for each country has identified the significant factors related to the role of the principal and the librarian in creating an information literate school community. Cross-country comparisons have identified some common concerns, priorities, and beliefs of principals and teacher-librarians across a diverse range of educational contexts. This is where individual countries could learn from each other regarding programs and strategies that effectively support the development of information literacy in schools. The findings from the cross-country comparisons were reported at the Third Annual International Forum on Research in School Librarianship, at the 1999 Joint Conference of AASL and IASL in Birmingham, Alabama. That report is available at http://athene.mit.csu.edu.au/~lhay/iasl/ilsc.html.
This project has demonstrated the potential benefits as well as the potential problems in conducting collaborative research in teacher-librarianship on an international scale. It is anticipated that the project will contribute to the development and publication of an international set of guidelines for principals and teacher-librarians in developing effective information services and supporting information literacy programs in schools.
[ Back to Top]
The authors acknowledge with thanks the contributions of Geoff Fellows of the Internet Special Project Group (ISPG) at Charles Sturt University for his programming knowledge and general technical know how.
Both the National Association of Elementarty School Principals ( www.naesp.org) and the National Association of Secondary School Principals ( www.nassp.org) have comprehensive sites that are invaluable for indicating the common concerns of principals‘ associations. [ back]
The International Association of School Librarianship ( www.hi.is/~anne/iasl.html) maintains an extensive Web site that contains information about the Association as well as many useful school library media resources. [ back]
The International Federation of Library Associations and Institutions ( www.ifla.org) also has a Web site with resources pertaining to library and information science programs, services, and research worldwide. [ back]
A recently updated ERIC Digest on flexible scheduling is available from http://ericir.syr.edu/Virtual/InfoGuides/alpha_list/Flex-Sched06-98.html). [ back]
- ERIC also has a 1997 InfoGuide on the “Evaluation of School Library Media Programs” ( http://ericir.syr.edu/Virtual/InfoGuides/alpha_list/Evalschlib12_97.html). [ back]
LaRocque, Linda, and Dianne Oberg. 1990. Building bridges between the library and the principal’s office. In Proceedings of the 19th annual conference of the International Association of School Librarianship, Umea, Sweden. Kalamazoo, Mich.: IASL.
LaRocque, Linda, and Dianne Oberg. 1991. The principal’s role in a successful library program. The Canadian School Executive 11, no. 4: 17–21.
Hay, Lyn, and James Henri. 1995. Leadership for collaboration: Making vision work. Paper presented at the IFLA Conference, School Libraries Programme Session, Istanbul, Turkey. Available online at www.ifla.org/IV/ifla61/61-hayl.htm.
Henri, James, and Lyn Hay. 1996. The principal’s role in developing and supporting an information literate school community. In Beyond the horizon: Conference proceedings of the fourteenth biennial conference of the Australian School Library Association, 111–25. West Perth, Australia: ASLA.
Oberg, Dianne. 1996. Principal support: What does it mean to teacher-librarians? In Sustaining the vision: A collection of articles and papers on research in school librarianship in honor of Jean E. Lowrie, ed. Laurel A. Clyde, 221–30. Castle Rock, Colo.: Hi Willow Research. Available online at www.ualberta.ca/~doberg/prcsup.htm.
Dogg Hafsteinsdottir, Hafdis. 1997. The attitude of Icelandic principals towards school libraries: A survey. Paper presented at the IFLA Conference, Section of School Libraries and Resource Centres, Copenhagen, Denmark. Available online at www.ifla.org//IV/ifla63/63hafh.htm.
Henri, James, and Lyn Hay. 1997. Understanding principal patronage: Developing and piloting a quantitative instrument. Paper presented at the IFLA Conference, Section of School Libraries and Resource Centres, Copenhagen, Denmark. Available online at www.ifla.org/IV/ifla63/63heha.htm.
Moore, Penny. 1997. Teaching information problem solving in primary schools: An information literacy survey. Paper presented at the IFLA Conference, Section of School Libraries and Resource Centres, Copenhagen, Denmark. Available online at www.ifla.org/IV/ifla63/63moop.htm.
Oberg, Dianne. 1997. Principal support: Research from Canada. Paper presented at the IFLA Conference, Section of School Libraries and Resource Centres, Copenhagen, Denmark. Available online at www.ifla.org/IV/ifla63/63obed.htm.
Charrier, Colette, Brigitte Bacconnier, and Alain Van Cuyck. 1998. The France experience. [The role of the principal in an information literate school community: An international research panel]. Paper presented at the IFLA Conference, Section of School Libraries and Resource Centres, Amsterdam, Holland. Available online at http://farrer.riv.csu.edu.au/principal/survey/paper/FR_html.
Han, Yook-ok. 1998. Principals and school librarians working within an information literate school community-Research from Korea [The role of the principal in an information literate school community: An international research panel]. Paper presented at the IFLA Conference, Section of School Libraries and Resource Centres, Amsterdam, Holland. Available online at http://farrer.riv.csu.edu.au/principal/survey/paper/KR_html.
Hay, Lyn, James Henri, and Dianne Oberg. 1998. The principal’s role in developing information literacy: Findings from Australia and Canada. In Education for all: Culture, reading and information, ed. S. Shoham and M. Yitzhaki, 69–80. Proceedings of the 27th International Conference of the International Association of School Librarianship, Ramat-Gan, Israel. Ramat-Gan, Israel: Dept. of Information Studies and Librarianship, Bar-Ilan Univ.
Henri, James. 1998. Developing information literate schools: Findings from Australia [The role of the principal in an information literate school community: An international research panel]. Paper presented at the IFLA Conference, Section of School Libraries and Resource Centres, Amsterdam, Holland. Available online at http://farrer.riv.csu.edu.au/principal/survey/paper/AU_html.
Herring, James. 1998. The Scotland experience: A report [The role of the principal in an information literate school community: An international research panel]. Paper presented at the IFLA Conference, Section of School Libraries and Resource Centres, Amsterdam, Holland. Available online at http://farrer.riv.csu.edu.au/principal/survey/paper/SC_html.
Kiinikangas, Liisa. 1998. Principals and school librarians working within an information literate school community-case Finland [The role of the principal in an information literate school community: An international research panel]. Paper presented at the IFLA Conference, Section of School Libraries and Resource Centres, Amsterdam, Holland. Available online at http://farrer.riv.csu.edu.au/principal/survey/paper/FI_html.
Oberg, Dianne. 1998. The Canadian experience: Research report [The role of the principal in an information literate school community: An international research panel]. Paper presented at the IFLA Conference, Section of School Libraries and Resource Centres, Amsterdam, Holland. Available online at
Introduction and Table 2
Austrom, L., R. Kennard, J. Naslund, and P. Shields. 1989. Implementing change: A cooperative approach. Vancouver, BC: British Columbia Teacher-Librarians’ Association.
Baker, D. P. 1980. The library needs the principal, too.
Instructor 89, no. 9: 125.
Browne, K. S., and L. Burton. 1989. Timing is everything: Adapting to the flexible schedule.
School Library Journal 34 (Dec.): 20–23.
Carson, B. B. 1989. Interacting for excellence. In
School Library Annual, Vol. 7, ed. J. B. Smith, 86–95. Englewood, Colo.: Libraries Unlimited.
Charter, J. B. 1982. Case study profiles of six exemplary public high school library media programs (Doctoral dissertation, Florida State University, 1982).
Dissertation Abstracts International 43 293A–294A.
Corr, G. P. 1979. Factors that affect the school library media specialist’s involvement in curriculum planning and implementation in small high schools in Oregon (Doctoral dissertation, University of Oregon, 1979). Dissertation Abstracts International 40 2955A.
Davies, R. A. 1979. The school library media program: Instructional force for excellence. 3d ed. New York: Bowker.
Dekker, B. 1989. Principals and teacher-librarians—Their roles and attitudes regarding school libraries: Results of a survey of elementary schools in Ontario. School Libraries in Canada 10, no. 2: 32–37.
Fox, M. 1982. The administrator and the school library. The Bookmark 24, no. 2: 11–19.
Grant, M. A. 1988. The principal’s role in the achievement of learning through the library media center. Catholic Library World 60: 71–74.
Hamilton, D. 1983. The principal and the school library. Education Canada 23, no. 3: 30–38.
Hellene, D. L. I. 1974. The relationships of the behaviors of principals in the state of Washington to the development of school library/media programs (Doctoral dissertation, University of Washington, 1973). Dissertation Abstracts International, 34, 3835A.
Kuehn, P. R. 1975. The principal and the media center. NASSP Bulletin 59, no. 392: 51–60.
Loertscher, D. D. 1988. Taxonomies of the school library media program. Englewood, Colo.: Libraries Unlimited.
Lundin, R. 1983. The teacher-librarian and information skills—An across the curriculum approach. Emergency Librarian 11, no. 1: 8–12.
Morris, B. J., J. T. Gillespie, and D. L. Spirt. 1992. Administering the school library media center. 3d ed. New York: Bowker.
Podemski, R. S. 1990. The principal: Key to role effectiveness. In School Library Media Annual, Vol. 8, ed. J. B. Smith, 24–31. Englewood, Colo.: Libraries Unlimited.
Prostano, E. T., and J. S. Prostano. 1987. The school library media center. 4th ed. Littleton, Colo.: Libraries Unlimited.
Shields, P. 1987. Teacher-librarians: Educational change agents (Master’s project, University of British Columbia, 1987).
Turner, P. M. 1980. The relationship between the principal’s attitude and the amount and type of instructional development performed by the media professional. International Journal of Instructional Media 7, no. 2: 127–38.
Wilson, P. J., M. Blake, and J. A. Lyders. 1993. Principals and teacher-librarians: A study and a plan for partnership. Emergency Librarian 21, no. 1: 18–24.
Yesner, B. L., and H. L. Jay. 1987. The school administrator’s guide to evaluating library media programs. Hamden, Conn.: Library Professional Publications.
Yetter, C. L. 1994. Resource-based learning in the information age school: The intersection of roles and relationships of the school library media specialist, teachers, and principal (Doctoral dissertation, Seattle University, 1994). Dissertation Abstracts International 55: 1130A.
Manuscript submitted: September 1998
Revised: April 1999, January 2000
Board approved: April 2000