Factors Influencing Information Literacy of University Students

Abstract

During the COVID-19 pandemic, effective use of information and communication technology (ICT), access to data sources, and critical evaluation of new information were essential for successful distance learning. University students need both information literacy (IL) and scientific literacy (SL) to learn and conduct research. This study examined the level of IL of 561 undergraduate and graduate students. We investigated the impact of scientific literacy (SL), ICT use, psychological/learning characteristics, and demographic parameters on student IL. The effects of a credit-bearing IL course were studied on 151 students, comparing three teaching methods. The average IL test performance of 67.6% did not differ significantly by student gender or natural/social science orientation. Of the IL topics, students were least proficient in legal/ethical issues, followed by information searching. Students’ knowledge of IL and SL was comparable and decreased with cognitive level. While ownership of ICT devices and ICT-rich courses had no effect on the level of IL, confidence in using the Internet correlated significantly with IL. Also, IL correlated positively with students’ self-concepts about learning and problem-solving, as well as their self-efficacy, but motivation played a smaller role. The credit-bearing IL study course was most effective when active learning methods were used.

1. Introduction

The COVID-19 pandemic brought many challenges, including those in the field of education, where most learning was switched to the online format almost overnight [1, 2]. Moreover, the COVID -19 information crisis was indicative of the more general problem of information overload in academic research. To improve information retrieval capabilities, students and researchers needed to improve their information retrieval skills and the systems they used [3].

The success of transition to online learning has been conditioned by multiple factors. Adequate access to information and communication technology (ICT) for both students and teachers was the first prerequisite to embark on online learning [4]. That could be hindered by slow/intermittent internet connections and incompatible or outdated devices and software.

The next requirement was related to proficiency of ICT use. A collective of skills, knowledge, and attitudes, labeled as digital competence, enabled students to effectively, efficiently, and ethically collaborate, solve problems, and manage information [5]. The Digital Competence Framework for Citizens (DigComp 2.2) provides a common understanding of what digital competence is; gives examples of knowledge, skills, and attitudes that help citizens engage confidently, critically, and safely with digital technologies; and proposes that the framework be modeled after the Digital Accessibility Guidelines [6]. Today’s students, the generation of so-called digital natives due to being born in the digital age, are expected to be digitally competent and handle ICT tools and applications in a natural way [7, 8]. However, that is not always the case, as some studies show [9].

In addition to access to ICT and digital literacy, other skills are crucial to navigating the vast online information landscape: knowing how to find, evaluate, process, and use information. Those are some key characteristics of the information literacy (IL). While the ICT and digital literacies focus primarily on skills associated with various digital technologies, IL is defined as an intellectual framework for understanding, finding, evaluating, and using information [10]. Different frameworks and sets of standards of IL are in use in various countries and at various education levels. Some of the most known standards and frameworks to be applied at the university level are shown in Table 1.

Year	Author/Institution	Framework name	Source
1990	Eisenberg & Berkovitz	Big Six	[11]
1997	Bruce	Seven Faces of Information Literacy in Higher Education	[12]
2000	ACRL	Information Literacy Competency Standards for Higher Education	[10]
2002	JISC	The Big Blue report: information skills for students	[13]
2004	Bundy/ANZIIL	Australian and New Zealand information literacy framework	[14]
2008	Catts & Lau/UNESCO	Six Skills	[15]
2011	SCONUL	Seven Pillars of Information Skills	[16]
2016	ACRL	Framework for Information Literacy for Higher Education	[17]

Table 1.

List of IL standards and frameworks.

At the university level in Slovenia, the ACRL standards/framework were adopted and translated into Slovenian language. At the basic level, the standards define IL by describing five key characteristics of an information literate student, who should be able to:

• determine the extent of information needed;

• access the needed information effectively and efficiently;

• evaluate information and its sources critically, and incorporate selected information into one’s knowledge base and value system;

• use information effectively to accomplish a specific purpose; and

• understand the economic, legal, and social issues associated with the use of information, and access and use information ethically and legally.

Studies have shown that during the COVID-19 pandemic, IL had a positive effect on students’ intention to use digital technologies for learning, performance expectancy, effort expectancy, habit, and hedonic motivation [18]. IL was critical not only for students but also for educators. There was a recognized need for more IL instruction for students and teachers [19, 20]. Learning success also depends on the teaching methods. Active learning methods in teaching IL were previously developed both for an online setting [21] and for large enrolment courses [22]. The appropriate use of technology for a chosen method plays a crucial role, and the applicability goes beyond the COVID-19 era.

IL and related skills are important both for the students involved in the formal learning process as well as in the daily lives of informed and responsible citizens. Studies show that digital natives are not automatically information literate [23]. Individuals with a lower level of IL, who do not possess the ability to critically evaluate information, are more susceptible to misinformation and fake news, for example, on the topics of climate change and vaccine safety. A study [24] reported that information literacy, which emphasized users’ ability to find verified and reliable information, was positively associated with fake news identification, while digital and media literacy showed no significant relationship. During the COVID-19 pandemic, the harmful consequences of spreading misinformation due to insufficient levels of IL became even more evident than in the past [25, 26].

Not only was the ability to judge the veracity of information by its source vital but also was the ability to find reliable and verified scientific information, accomplished with suitable information searching skills and access to credible information sources [27]. Scientific databases, where most factual information can be found, are usually subject to copyright restrictions and are not freely available to citizens, and sometimes, this even holds true for mainstream media that are tasked with informing the public [28]. University students usually have licensed access to reliable scientific databases. However, many students view the process of searching for information as laborious [29]. It is therefore critical that students be supported in developing information literacy skills, including the use of reliable scientific databases with advanced search techniques. Research [3] has established that the search skills require dedicated education and training for all three main types of searches that researchers perform: lookup searches conducted with a clear goal in mind; exploratory searches to better understand the nature of a topic; and systematic searching with the goal to identify all relevant information sources in a transparent and reproducible manner. These three types must be performed with different search methods, using search systems with specific functionalities.

IL is not a closed set of abilities, but it is related to other abilities and characteristics of students. A close connection exists between IL and digital literacy, as represented in DigComp framework [6] with five competence areas, which combine elements of ICT literacy and IL. There are also parallels between scientific literacy and IL [30]. Some studies have investigated the factors that can influence students’ IL. A study by [31] found that the student’s IL was significantly influenced by both individual subjective factors, such as information processing learning style, and external objective factors, such as social media content consumption and content creation behaviors. While no significant difference in the level of IL was found between genders, IL differed significantly between fields of study and between students with different levels of academic achievement. In contrast, when self-reporting, boys tend to overestimate their ICT literacies, whereas girls appear to underestimate their capabilities [32]. One study [33] examined the relationship between IL and social media competence. The results showed that university students’ IL and ability to use information technology to solve problems, as well as their sense of responsible behavior in cyberspace, were the most important factors in predicting students’ social media competence. The implication is that enhancing university students’ IL will have a positive impact on university students’ social media behavior. Other research [34] examined how two emotional constructs (emotional intelligence and dispositional affect) and two cognitive constructs (motivation and coping skills) were related to students’ IL. The results of correlation and regression analyses showed that emotional intelligence and motivation significantly predicted students’ IL outcomes. Another study [35] studied the predictors of medical students’ IL self-efficacy skills. Results suggested that emotional intelligence subconstructs (appraising own emotions, appraising others’ emotions, and using emotions) had a statistically significant positive impact on students’ IL self-efficacy.

2. Aims and scope of the study

In this chapter, we present and discuss the IL as measured in a group of 561 undergraduate and graduate university students, aiming to answer the following research questions:

RQ1: What is the level of IL among students? How is it affected by demographic parameters, such as gender, type of study major, and study year?

RQ2: In which content areas of IL are students successful, and in which areas should they be given more emphasis in their education?

RQ3: Is there a relationship between students’ IL and their scientific literacy? Are students’ abilities to master higher levels of cognition (understanding and applying knowledge) comparable between the two literacies?

RQ4: Does software use, ownership of ICT devices, number of ICT-rich courses, and confidence in using the Internet affect students’ level of IL?

RQ5: How is IL influenced by various psychological/learning parameters, such as self-concept about learning and problem-solving, general self-efficacy, use of metacognitive learning strategies, internal motivation, and autonomous and controlled external motivation?

RQ6: To what extent does a study course with IL content contribute to improving students’ information literacy? How do the teaching methods affect the outcomes?

RQ7: How much of the IL could be explained by demographic parameters, scientific literacy, ICT use, and psychological/learning parameters of students? Which parameters affect IL levels the most?

3. Methods

3.1 Research instruments

Four tests and questionnaires were applied in our study: an IL test, a scientific literacy test, a questionnaire on ICT use, and a questionnaire on psychological/learning leanings of participants. Additionally, a teaching intervention, namely a dedicated IL course, was implemented, using three different teaching methods.

3.1.1 Information literacy test (ILT)

A multiple-choice knowledge test [36] was used, comprising 40 multiple choice items with four options and one correct, yielding a point per item. For analysis purposes, ILT items were divided into subscales by five ACRL 2000 information literacy standards (A1—information needs identification, A2—information search, A3—information evaluation, A4—information use, A5—ethical/legal issues) [10]. Similarly, ILT items were classified into one of the three cognitive categories (B1—remembering, B2—understanding, B3—applying), simplified from the Bloom’s Taxonomy [37].

3.1.2 Scientific literacy test (SLT)

A mixed-type knowledge test was applied, consisting of six problem-based tasks related to popular science topics, totaling 23 items with as many points. The problems were selected from the PISA 2006 science survey [38]. While some of the items were multiple-choice, others were open ended and had to be evaluated manually. SLT items were also assigned one of the three cognitive categories previously described (B1—remembering, B2—understanding, B3—applying).

3.1.3 Questionnaire on ICT use

We used a 35-item scale with four subscales. The first two aimed to measure software (ICT-S, 16 items) and hardware use (ICT-H, 4 items) on a 5-point Likert scale, reflecting frequency of use (never, less than once a week, multiple times a week, almost every day, multiple times a day). The third subscale (ICT-C, 5 items) inquired about the number of ICT-rich study courses students were enrolled in. Confidence of Internet use (ICT-I, 10 items) was surveyed in the fourth segment on a 5-point Likert scale, based on the degree of agreement with given statements.

3.1.4 Questionnaire on psychological/learning factors

A 70-item questionnaire on a 5-point Likert scale (based on the agreement level) was utilized to measure components of psychological/learning leanings of individuals [39]. Questionnaire items were compiled from Self-description questionnaire III (SDQ , [40]), Generalized self-efficacy scale (GSE, [41]), and the Academic motivation questionnaire [42]. The seven subscales applied were self-concept about learning (SC-L, 10 items), self-concept about problem-solving (SC-P, 10 items), self-efficacy (SE, 10 items), use of metacognitive learning strategies (LS, 15 items), internal motivation (IM, 13 items), autonomous external motivation (EM-A, 6 items), and controlled external motivation (EM-C, 6 items).

Reliability of the four instruments, exhibited as Cronbach α on the testing sample, is shown in Table 2.

Scale	Description	Items	Cronbach α
Literacy test
IL	Information literacy	40	0.724
SL	Scientific literacy	23	0.608
ICT use
ICT-S	Software use	16	0.728
ICT-H	Hardware possession	4	0.411
ICT-C	Number of ICT rich courses	5	0.667
ICT-I	Internet confidence	10	0.822
Psychological/learning factors
SC-L	Self-concept about learning	10	0.805
SC-P	Self-concept about problem-solving	10	0.765
SE	Self-efficacy	10	0.853
LS	Metacognitive learning strategies	15	0.683
IM	Internal motivation	13	0.844
EM-A	Autonomous external motivation	6	0.716
EM-C	Controlled external motivation	6	0.653

Table 2.

Size and reliability of research instruments.

3.1.5 Dedicated IL course

Impact of the study course with 45 contact hours, conducted in one semester and bearing 3 credit points, was also explored in this research. The course content was in line with the five ACRL IL standards [10]: information need identification, information search, information evaluation, information use, and legal/ethical issues. Three different teaching methods were applied in the course.

• In the lecture-based group, traditional lectures were given, following the sequence of chapters from the curriculum. In the hands-on computer lab, students worked on predefined database search exercises, but on individual topics.

• In the project-based learning group, students worked on individual project topics with the goal of producing a review article. They went through the research steps of specifying information need, formulating queries and database searching, and evaluating and synthesizing search results. Lectures were given as organized support for project work.

• In the problem-based group, students had the goal of solving a selected complex problem from their field of study. The main problem was broken down into subproblems. Lectures were delivered mainly as directed interventions, explanations, instructions, or answers to students’ questions to facilitate the problem-solving process. Students worked in small groups, and the hands-on work followed the tasks and group dynamics in solving the problem. Final reports were presented and discussed in the form of student conference.

3.2 Participants and procedure

The testing group comprised 561 university students of two universities and one independent higher education school in Slovenia. Composition of the group by gender, study year, and type of study major is shown in Table 3.

Parameter	Value	Students	%
Gender	Male	190	33.9
	Female	371	66.1
Type of study major	Natural sciences	397	70.8
	Social sciences	164	29.2
Study year	1	291	51.9
	2	125	22.3
	3–4	82	14.6
	5–6	63	11.2

Table 3.

Testing group composition by demographic parameters.

All 561 students took the IL test before taking any IL-dedicated classes. At the same time, they also took the SL test and both ICT and psychological questionnaires. Of the 561 students, 151 later took a credit-bearing IL course, described in the instruments section. This group of students took the IL test again as a post-test, so that the change in their IL skills could be studied in comparison to the test results before the course (pre-test). The 151 students were divided into three groups based on the teaching method used in the course: lecture-based (52 students), project-based (52 students), and problem-based (47 students) learning.

The online survey system 1 ka (1 ka.si) was applied for testing, which took place at university locations, in presence of a professor. Before testing, an introductory protocol was administered, providing explanation of the study goals and assurances of anonymity and voluntary, emphasizing participation. There was no time limit for completing the tests and questionnaires.

3.3 Analyses

Reliability in terms of Cronbach alpha was calculated for IL and SL tests, 4 ICT, and 7 psychological/learning subscales. IL score means were analyzed, both total scores and partial scores, corresponding to the five IL content categories and three cognitive categories. Differences in IL levels between the pre-test and the post-test were measured with paired t-tests. Differences in IL between teaching methods and differences in IL between demographic parameters (gender, type of study major, study year) were investigated using two-sample t-tests. SL score means were analyzed for total scores and cognitive categories. Means were also calculated for ICT subscales as well as psychological subscales. Pearson’s correlation coefficients were calculated between IL and SL, their content/cognitive subscales, and ICT and psychological subscales. Multiple linear regression was applied to predict the IL level from other parameters: SL, demographic parameters, ICT, and psychological/learning parameters. All data collected were analyzed using Microsoft Excel.

4. Results and discussion

4.1 IL and demographics

RQ1: What is the level of IL among students? How is it affected by demographic parameters, such as gender, type of study major, and study year?

In the ILT test, students achieved a moderately high mean level of IL at 67.63% (Table 4). The lowest value was 20% and the highest 100%. Histogram shows a normal-like distribution of ILT scores (Figure 1).

Statistic	Value (%)
Mean	67.63
Standard error	0.52
Median	67.50
Mode	65
Standard deviation	12.37
Sample variance	153.10
Kurtosis	0.41
Skewness	−0.48
Range	80
Minimum	20
Maximum	100
Count	561
Confidence Level (95.0%)	1.0261

Table 4.

Descriptive statistics of the ILT score (N = 561).

Figure 1.

ILT score distribution (N = 561).

The influence of demographic parameters on ILT scores was studied next (Figure 2). No statistically significant difference in IL level was found between male and female students or between the main two types of study majors (natural sciences and social sciences). There was only a 1% difference between groups in both demographic categories. Significant difference was observed only between the year 2 and years 3–4 students (6% difference). This could indicate that students did not have enough opportunity to practice their IL skills in the first two years, probably due to the curriculum with the traditional basic courses, while in the later years, with the specific elective courses, students were more likely to develop the IL skills due to the nature of the assignments and active learning. The lack of a difference between natural science and social science majors could mean that both groups had courses in their curricula that facilitated the development of IL. Pearson’s correlation between study year and IL was 0.14, which is statistically significant but small. The negligible improvement in IL between academic years 1 and 2 and the dominance of general courses in year 1 suggest that more IL needs to be introduced in the second year, whether through a special IL course or through the existing courses.

Figure 2.

ILT score means according to demographic parameters (M—male, F—female; sci—natural sciences, non-sci—social sciences; N = 561).

4.2 IL content

RQ2: In which content areas of IL are students successful, and in which areas should they be given more emphasis in their education?

Partial ILT scores based on content categories were investigated (Figure 3). The lowest mean was achieved in the content category of legal and ethical use of information (A5—55%), followed by information search (A2—65%) and information need identification (A1—69%). Students were more successful in information use (A4—73%) but especially in information evaluation (A3—83%). These results suggest that during IL courses, more emphasis should be put on ethical, legal, and socio-economic aspects of information use, as well as on advanced database searching techniques. Lack of information searching skills can hinder students’ research work, on one hand, as well as affects citizens’ ability to verify information when confronted with dubious claims either in social media or in other information sources that may seem legitimate at first glance. On the other hand, the high level of competence in evaluating information may indicate that university students are not as susceptible to deliberate misinformation as the general population and that students are relatively good at applying criteria to evaluate the credibility of information sources.

Figure 3.

ILT score means according to IL content categories (whiskers represent SD; N = 561).

Achievements in different IL content categories were interconnected with Pearson’s correlations among categories ranging from 0.21 to 0.39. The lowest value was achieved between information use and the two weakest IL categories, namely information search and ethical issues.

When proficiency in individual IL content categories was studied in light of demographics, it turned out that female students performed significantly better in information evaluation than males, while social science majors performed better in ethical issues than natural science majors. The biggest difference among lower and higher year students was achieved in ethical issues (8% difference) and information use (10% difference), but students were closer in information evaluation (3% difference) and information search (5%).

4.3 IL and scientific literacy

RQ3: Is there a relationship between students’ IL and their scientific literacy? Are students’ abilities to master higher levels of cognition (understanding and applying knowledge) comparable between the two literacies?

On the SLT knowledge test, students achieved very similar total proficiency levels to the ILT test (mean 67.63% on ILT vs. 67.02% on SLT). Score distribution was similar as well. The two scores correlated significantly, with Pearson’s correlation of 0.44.

IL and SL scores were evaluated on a cognitive subscale. Results showed a similar level of IL on the cognitive level of remembering (B1) and understanding (B2), but students were less successful in knowledge application (B3) (Figure 4). Their SL proficiency decreased more with each cognitive level.

Figure 4.

Comparison of ILT and SLT score means according to cognitive categories (whiskers represent SD; N = 561).

In terms of demographics, no differences were observed between genders in IL or SL. As expected, natural science students were significantly better in SL than social science students, (understanding—B2; 72% vs. 66%, and applying—B3; 48% vs. 43%). Differences among study years 2 and 3–4 were observed mostly in the lowest two cognitive categories, but not in knowledge application.

4.4 IL and ICT

RQ4: Does software use, ownership of ICT devices, number of ICT-rich courses, and confidence using the Internet affect students’ level of IL?

Results showed no correlation between ILT score and device ownership (ICT-H, Table 5), nor between ILT and number of ICT-rich study courses (ICT-C). Correlation with software use (ICT-S) was slightly higher, but the highest and statistically significant correlation was with confidence using the internet (ICT-I), reaching 0.19.

Parameter	ICT-S	ICT-H	ICT-C	ICT-I
Pearson’s r	0.08	0.01	−0.04	0.19

Table 5.

Pearson’s correlation of ICT scales with ILT score (N = 561).

ICT parameters were interrelated: software use correlated with hardware possession and internet confidence (r = 0.38), while hardware possession also correlated with internet confidence (r = 0.19).

Analysis by demographic parameters showed that male students used software more often, possessed more devices, and were more confident using the internet than females, despite female students taking more ICT-rich courses. Social sciences majors used software more often, possessed more devices, and were more confident using the internet than natural sciences majors. Significant increase in software use and internet confidence was observed from year 2 to years 3–4.

4.5 IL and psychological leanings

RQ5: How is IL influenced by various psychological/learning parameters, such as self-concept about learning and problem-solving, general self-efficacy, use of metacognitive learning strategies, internal motivation, and autonomous and controlled external motivation?

Three of the seven psychological scales correlated significantly with the ILT score (Table 6): self-concept about learning (SC-L) and problem-solving (SC-P) as well as self-efficacy (SE). This result was expected as the abilities to learn, solve problems, and be efficient are more likely to lead to success. With low correlation, the use of metacognitive learning strategies (LS) was not found as an important factor. When students did not understand the material, they asked their classmates rather than a teacher for help.

Parameter	SC-L	SC-P	SE	LS	IM	EM-A	EM-C
Pearson’s r	0.26	0.24	0.22	0.07	0.12	0.11	−0.08

Table 6.

Pearson’s correlation of psychological/learning scales with ILT score (N = 561).

Motivation played a smaller role (Table 6). Internal motivation (IM) and autonomous external motivation (EM-A) correlated only slightly with the ILT score. Internally motivated students highly rated their interest in and understanding of the field of study. In the external autonomous scale, students most acknowledged the value of learning for their future—their employment prospects and professional development. Controlled external motivation (EM-C) did not correlate well with ILT. Item analysis of this scale showed that most students did not consider it important to make an impression on the teacher, parents, or their peers but that they nevertheless relied on the teacher’s authority and were motivated by good grades.

All psychological parameters correlated rather heavily among themselves. For example, the strongest link was found between self-concept about learning and internal motivation, and another link between self-concept about problem-solving and self-efficacy (both r = 0.67). In other correlations related to IL, three of the psychological parameters correlated significantly to student confidence using the internet (scale ICT-I), namely self-concept about problem-solving (r = 0.29), self-efficacy (r = 0.31), and internal motivation (r = 33).

Regarding demographic parameters, female students scored significantly higher in self-concept about learning and using metacognitive learning strategies, while male students scored higher in self-concept about problem-solving. The type of study major played no role in the psychological parameters, but year of study did in all, except in problem-solving. The problem-solving ability seems to be a personal characteristic rather than an acquired skill.

4.6 IL study course and teaching methods

RQ6: To what extent does a study course with IL content contribute to improving students’ information literacy? How do the teaching methods affect the outcomes?

Results are shown for the subgroup of 151 students who were enrolled in an information literacy course and took the ILT test before (pre-test) and after the course (post-test). Students’ mean IL level improved significantly from 65% on the pre-test to 80% on the post-test (Figure 5). Significant improvement was achieved in all IL content categories, but it was the highest in information use (A4—25%), information search (A2—19%), and ethical issues (A5—17%). The lowest increase was observed in information evaluation (A3—10%) due to the fact that the pre-test level was already high. When it came to cognitive categories, the largest increase was obtained in the highest category of applying (B3—22%) and the lowest in understanding (B2—11%).

Figure 5.

Comparison of ILT scores on the pre-test and post-test for total IL, five content and three cognitive categories (N = 151).

Students were divided into three groups, based on the teaching method applied in the course: traditional lectures (52 students), project-based learning (52 students), and problem-based learning (47 students). Improvement in total IL according to the teaching method is shown in Figure 6. The biggest improvement was achieved in project-based learning group (PRJ—18%), followed by the problem-based group (PBL—16%). The traditional lecture group (LEC) improved for 11%, suggesting that the active teaching methods were more successful and could be recommended for university IL study courses.

Figure 6.

ILT score means for three teaching methods (N_LEC = 52, N_PRJ = 52, N_PBL = 47) on the pre-test and post-test.

Figure 7 shows the pre-test scores and progress achieved on the post-test for each of the three teaching methods in every IL content categories. The biggest improvement for all three teaching methods was achieved in the IL category of information use (A4; 15–42%), followed by information search (A2; 13–26%) and legal/ethical issues (A5; 13–18%). The lowest progress was obtained in the category of information evaluation (A3; 5–13%) and identification of information needs (A1; 10–15%). Both groups using active teaching methods (project- and problem-based learning) achieved similar post-test proficiency in all five content categories, which was above that of the traditional lecture group.

Figure 7.

ILT scores on the pre-test and the progress achieved on the post-test for three teaching methods (N_LEC = 52, N_PRJ = 52, N_PBL = 47) and IL content categories.

Another look at improvements across the cognitive categories (Figure 8) shows the highest increase for all three teaching methods in the highest category of applying knowledge (B3; 19–27%), followed by the lowest category of remembering (B1; 10–17%), while improvement was the lowest in the category of understanding (B2; 6–15%). The two active learning methods outperformed the lecture-based approach in most cases.

Figure 8.

ILT scores on the pre-test and the progress achieved on the post-test for teaching methods (N_LEC = 52, N_PRJ = 52, N_PBL = 47) and IL cognitive categories.

4.7 Factors influencing IL

RQ7: How much of the IL could be explained by demographic parameters, scientific literacy, ICT use, and psychological/learning parameters of students? Which parameters affect IL level the most?

In an attempt to develop a model for prediction of the IL level, taking into account the following independent variables (3 demographic parameters, SL, 4 ICT scales, and 7 psychological/learning subscales), we applied multiple linear regression on the test results of all participants (561). The model accounted for 29% of the variance in IL; F(15, 545) = 15.06, p < 0.001, R² = 0.293. As shown in Table 7, predictors with a significant zero-order correlation with IL (scientific literacy SL, confidence on the internet ICT-I, and self-concept about learning SC-L) had a significant partial effect in the full model. Contributions to IL variance, calculated by partitioning R² by multiplying beta values with zero-order correlations, were 16.6% (SL), 2.8% (ICT-I), and 3.5% (SC-L).

	Unstand. coeff.		Stand. coeff.			95.0% Confid. Inter. for B		Correlations
Predictor	B	Std. err.	beta	t	Sig.	Lower	Upper	Zero-; order	Partial
(Constant)	0.177	0.065		2.740	0.006	0.050	0.304
Gender	0.012	0.012	0.046	1.012	0.312	−0.011	0.036	0.017	0.043
Major	0.005	0.012	0.017	0.379	0.705	−0.020	0.029	−0.011	0.016
Year	0.005	0.003	0.059	1.433	0.153	−0.002	0.012	0.137	0.061
SL	0.363	0.034	0.402	10.579	0.000	0.296	0.431	0.445	0.413
ICT-S	−0.003	0.013	−0.009	−0.219	0.826	−0.029	0.023	0.085	−0.009
ICT-H	0.006	0.009	0.028	0.682	0.495	−0.011	0.023	0.013	0.029
ICT-C	−0.005	0.002	−0.081	−2.148	0.032	−0.009	0.000	−0.035	−0.092
ICT-I	0.033	0.008	0.166	3.969	0.000	0.017	0.049	0.194	0.168
SC-L	0.048	0.012	0.208	3.971	0.000	0.024	0.072	0.261	0.168
SC-P	0.020	0.012	0.087	1.695	0.091	−0.003	0.043	0.244	0.072
SE	0.011	0.013	0.047	0.841	0.401	−0.014	0.036	0.220	0.036
LS	−0.029	0.0140	−0.104	−2.089	0.037	−0.057	−0.002	0.072	−0.089
IM	−0.026	0.015	−0.115	−1.819	0.069	−0.055	0.002	0.122	−0.078
EM-A	0.010	0.012	0.044	0.834	0.404	−0.013	0.033	0.106	0.036
EM-C	−0.005	0.008	−0.025	−0.632	0.528	−0.020	0.010	−0.081	−0.027

Table 7.

Predictors of IL (statistically significant predictors are bolded; N = 561).

Number of ICT-rich courses had a significant negative effect in the model, but since its zero-order correlation with IL was 0, it played a suppressor role, strengthening the effect of other parameters. The application of metacognitive learning strategies (LS) also had a negative influence in the model, which was in line with its small (albeit non-significant) correlation with the ILT.

5. Conclusions

Previous studies have shown that during the COVID-19 pandemic, students’ access to information and communication technologies, ICT skills, and IL were critical to their shift to online distance education and to overcoming multiple chaotic information problems.

In this study, we took a closer look at university students’ IL and investigated the factors affecting students’ IL. Based on the results measured in a group of 561 students using IL and SL tests as well as questionnaires on students’ ICT use and psychological/learning characteristics, we came to the following conclusions:

RQ1. According to the results of the IL test, students are reasonably well information literate, and the IL does not differ by student gender or natural/social science orientation.

• Students achieved a moderately high mean level of IL, 67.63%, with a normal-like distribution of scores.

• No statistically significant difference in IL level was found between male and female students or between natural science and social science majoring students.

RQ2. Students are not equally skilled in all content areas of IL.

• Students were most successful in information evaluation (83%) and information use (73%).

• Most IL deficits were found in legal and ethical use of information (55%) followed by information search (65%). Therefore, more emphasis should be placed on those topics in higher education.

RQ3. There is a relationship between IL and students’ scientific literacy. In both areas, students have comparable skills and achieved similar results on three cognitive levels.

• Students scored similarly on the IL and SL tests, with a similar distribution of total scores and similar performance on the cognitive levels of remembering, understanding, and knowledge application.

• Females and males were equally successful - no gender differences were found in IL or SL test scores.

RQ4. Ownership of ICT devices and ICT-rich courses do not necessarily lead to higher levels of IL among university students. However, there is a significant correlation between IL and students’ confidence using the internet.

• The highest statistically significant correlation was found between IL and confidence using the internet; a lower correlation was found between IL and use of software, while no correlation was found between ILT score and ownership of ICT devices or number of ICT-rich courses.

• Male students owned more ICT devices, used software more often, and were more confident in using the internet than female students.

RQ5. Information literacy is influenced by some psychological parameters.

• Significant correlations were found between IL and self-concept of learning, self-concept about problem-solving, and self-efficacy.

• Motivation played a minor role; internal motivation and autonomous external motivation correlated only slightly with IL test scores, while controlled external motivation did not correlate at all with IL scores.

RQ6. An efficient way to reach a higher level of students’ IL is to introduce a credit-bearing study course that covers all the major subject areas of IL, preferably with the use of active teaching methods.

• The study course significantly contributed to a higher level of students’ IL (the average IL level improved from 65% on the pre-test to 80% on the post-test).

• Improvement was made in all IL content categories but most notably in the areas of information use, information search, and ethical issues, where students initially had the most difficulty.

• All three teaching methods (lecture-based, project-based, and problem-based learning) were successful in teaching IL.

• Active teaching methods slightly outperformed the traditional lecture-based approach.

• The positive contribution of active learning was the greatest in the knowledge application cognitive category.

RQ7. Students’ IL can be partially explained by scientific literacy, ICT, and psychological parameters.

• Of the 15 potential predictors of IL, which were included in a model to predict the level of IL, students’ scientific literacy, their confidence in using the internet, and their self-concept about learning had a significant effect.

• The model explained 29% of the variance in IL.