Developing and evaluating the feasibility of administering the SLUMP questionnaire for evaluating ergonomic exposures to laptop use in university students

Abstract

BACKGROUND:

Laptop computer use may lead to the development of health problems. However, few methods exist to measure laptop use in undergraduate students.

OBJECTIVES:

To develop a web-based questionnaire to measure ergonomic exposures to laptop use and to test the feasibility of administration in university students.

METHODS:

We pilot tested the administration of the Student Laptop Use and Musculoskeletal Posture (SLUMP) questionnaire to 44 second year undergraduate students. We measured the participation rate, completion of questions, and sought feedback for improving the questionnaire.

RESULTS:

The participation rate was 75% (33/44). Students needed 10–12 minutes to complete the questionnaire. Participants suggested adding half hour intervals for questions measuring breaks and improving the clarity by emphasizing the sections focusing on academic, recreational and employment laptop use.

CONCLUSION:

Our pilot study suggests that the SLUMP questionnaire is appropriate for “in-class” measurement of ergonomic exposures to laptop use in university students.

Keywords

Questionnaire development computer use undergraduate student

1 Introduction

University students rely on laptop computers to complete their education. In 2005, 52.8% of undergraduate students used a laptop for school [1]. Today, 95.4% of university students use a laptop computer [2]. Although only 4.6% of students use a desktop computer, the majority of research on ergonomic exposures pertaining to students still focuses on desktop computers which are scarcely used by university students [2].

Cross-sectional studies suggest that laptop use by university students is associated with health problems such as musculoskeletal pain, computer vision syndrome, alexithymia and mental health outcomes [3 –8]. Although these cross-sectional studies cannot be used to infer that laptop use is a risk factor for poor health, they nevertheless raise the hypothesis that laptops may lead to health problems [3 –7].

Few methods are available to collect ergonomic exposure data for laptop use in university students. Most research relies on biomechanical and ergonomic assessments [9 –12]. Measures such as muscle activation levels and efficiencies, joint angles, and passive motion analyses are often utilized to measure ergonomic exposure to laptop use [9 , 13]. While these assessments are reliable and valid, they are time consuming and impractical in epidemiological research. Questionnaires are the most effective way to measure laptop use for large sample sizes.

Several questionnaires are available to measure computer use; however most of these questionnaires aim to measure occupational use and/or desktop computers [14 –18]. There are a number of ergonomic differences between laptop and computer use, which are not addressed in these questionnaires. This is problematic because computer use may vary in university students compared to workers since 95.4% of students use laptops instead of desktop computers [2].

To our knowledge, three instruments were designed to measure laptop use in university students. The Boston University Computer and Health Survey was developed to evaluate musculoskeletal pain and activities associated with computing in university students [4, 19]. This web-based survey enquires about: 1) where students use their computer; 2) the proportion of time spent using a laptop and desktop computer; 3) the presence, severity and location of pain; and 4) activities associated with pain (laptop use, desktop computer use, or during other school related activities). The questionnaire focuses on the association between school— related activities and musculoskeletal pain. The reliability and validity of the questionnaire are unknown.

A second instrument, the Laptop Use Scale questionnaire was recently developed to measure student laptop use inside and outside classrooms [20]. The Laptop Use Scale aims to measure in-class academic use, in-class non-academic use, outside of class academic use, and outside of class non-academic use. However, this questionnaire does not address ergonomic exposures such as posture during laptop use. The Laptop Use Scale questionnaire has acceptable validity and reliability (Cronbach’s alpha = 0.77–87) [20].

The final instrument, the Student Laptop Use and Neck Pain Risk Questionnaire (SLUNPRQ) [21] was designed at the University of Ontario Institute of Technology (UOIT) in 2011. The items were selected based on a review of the literature and consultation with expert clinicians and researchers. The SLUNPRQ was developed to measure laptop use and its association with neck pain in university students (Questionnaire A) [21]. It includes questions regarding postures during recreational, academic and work-related laptop use. Breaks and external accessories are also evaluated in relation to laptop use for recreational, academic and work-related laptop use. The test-retest reliability of the questions in the SLUNPRQ ranged from k = 0.36–1.00 and the validity of the questionnaire has not been tested.

We modified the SLUNPRQ and adapted it to be administered online during class. The purpose of our study was to test the feasibility of administrating the new version of the questionnaire [renamed the Student Laptop Use and Musculoskeletal Posture (SLUMP) questionnaire] in undergraduate university students. Specifically, we aimed to determine the participation rate, completing time, and self-perceived clarity, length of the questionnaire and ease of navigation. We also aimed to collect preliminary data on laptop use to determine the frequency of outliers in the data.

2 Methods

2.1 Study design

We conducted a pilot study on September 25th 2015 to determine the feasibility of administering the web-based Student Laptop Use and Musculoskeletal Posture (SLUMP) questionnaire to undergraduate students (Questionnaire B).

2.2 Study sample

The study sample included 44 undergraduate students enrolled in a second-year anatomy course at the University of Ontario Institute of Technology (UOIT). UOIT was founded in 2003 in Oshawa, Ontario and has over 10,000 students enrolled. The university provides a technology enriched learning environment where all undergraduate students lease a laptop computer.

2.3 Development of the Student Laptop Use and Musculoskeletal Posture (SLUMP) Questionnaire

We modified and adapted the SLUNPRQ to develop the SLUMP questionnaire. We aimed to improve the clarity of questions by: 1) simplifying the language; 2) adding clear descriptions to figures demonstrating postures; and 3) ensuring that the response options truly reflect the reality of student life. Finally, the questionnaire was converted to an electronic format. The specific modifications are described below:

Modification #1: In the SLUNPRQ, all questions related to recreational, academic and work laptop use were asked in table formats (Questionnaire A, Q1— Q5). Most questions (22/ 33) had low to moderate test-retest reliability (k = 0.36 to k = 0.79). We attributed, in part, the lower than desirable reliability to the complexity of the table. We revised this section in the SLUMP questionnaire by separating each question individually and by providing separate sections for questions regarding academic, work and recreational use.

Modification #2: The SLUNPRQ response options for items inquiring about the duration of laptop use, postures while using a laptop, and duration and frequency of breaks were categorical in nature in the SLUNPRQ. These questions had low reliability (k = 0.37 to k = 0.69). Categorical responses may omit responses, which fall outside of the ranges provided. For example, students could only choose one posture that they use most frequently (Questionnaire A, Q2). Students who use multiple postures could not choose multiple responses. We revised the response options for these questions to include continuous variables in the SLUMP questionnaire (Questionnaire B, Q7— Q21). In the SLUMP, students may report how many hours per day they use each posture.

Modification #3: The SLUNPRQ allowed students to choose only one posture frequently used while using a laptop (Questionnaire A, Q2). The range in test-retest reliability for these questions (k = 0.36 to k = 0.80) may be due to the fact that it does not represent laptop use in university students; students likely use multiple postures while using a laptop. Therefore, in the SLUMP questionnaire, we revised these questions to allow students to select the number of hours per day in five different postures for recreational, academic or employment related purposes. Moreover, we divided these questions into weekdays and weekends since students may have different laptop habits during the school week when compared to weekends (Questionnaire B, Q17— Q21).

Modification #4: The SLUNPRQ used the term “work-related” when asking questions about laptop use for employment. The test retest reliability for questions related to posture during “work-related” laptop use ranged from k = 0.36 to k = 0.64. The use of the term “work-related” could have been misinterpreted by students to include school work. Instead, we replaced “work-related” with “employment-related” to distinguish between school work and job related laptop use in the SLUMP questionnaire.

Modification #5: The SLUNPRQ had a low test retest reliability ranging from k = 0.37 to k = 0.67 for questions regarding duration and frequency of breaks (Questionnaire A, Q5a, 5b). These questions did not provide a frame of reference for students to understand what were regarded as breaks. We revised these questions related to breaks in the SLUMP questionnaire to provide examples of what constitutes a break (Questionnaire B, Section 4). We also removed the question regarding how many breaks a person takes (Questionnaire A, Q5c). We replaced this question with “how often do you typically take breaks for recreational/ academic/ employment purposes? Every___hours” (Questionnaire B, Q25a, Q26a, Q27a). This provides participants with a frame of reference which may be easier to recollect.

Modification #6: The SLUNPRQ included questions to evaluate neck pain risk. We removed these items because there are multiple questionnaires with adequate psychometric properties which can be used to assess musculoskeletal pain.

Modification #7: The SLUNPRQ had several questions, which were not included in the analysis due to missing data. To minimize missing data, we converted the survey from a pen and paper questionnaire to a web-based questionnaire and required that each question be answered (students had to answer the question to move on to the next). Web-based questionnaires also allow for quick data collection, a larger sample size and easier navigation through questions with only a minor reduction in response rates [22, 23]. We also included skip logic in the questionnaire so students can avoid questions that do not apply to them. This should decrease the time to complete the questionnaire and reduce the burden on students.

The web-based version of the questionnaire was created using Google Forms and stored on a secure, password protected Google Drive account. The Google Drive account is hosted by the University of Ontario Institute of Technology and ensures that confidentiality is maintained.

Modification #8: The original survey instrument asked students to provide any additional reasons that they may experience neck pain however no results were reported. We replaced this question with an open-ended question where students could provide any additional comments regarding improvements to the questionnaire. We also included questions relating to the questionnaire design to gain additional feedback for future improvements (Questionnaire B Q28— Q30). These three questions requested students to use a 10-point scale to rate the questionnaire length and clarity and the ease of navigation.

2.4 Recruitment and data collection

The primary investigator (CD) received approval from the course director of a second-year anatomy course to recruit students from the class. The survey was conducted during a laboratory session; we selected to administer the questionnaire in class to facilitate participation and assist students with completion of the questionnaire. To minimize the risk of coercion, the professor and lab instructors were asked to leave the classroom during the administration of the questionnaire; but the primary investigator remained in the room to address any questions or technological problems with accessing or completing the questionnaire. At the beginning of the lab, the primary investigator addressed the class, informed students of the purpose of the study and discussed the data collection methodology. Following the presentation, the primary investigator read the informed consent form and allowed students to ask questions. Students were then asked to volunteer for the study.

Students who were interested in participating in the study were instructed to access the SLUMP Questionnaire online by using their school login to access the consent form and questionnaire. Students were instructed to remain seated if they chose not to participate. Students were informed that their participation was voluntary and that they could withdraw at any time by closing the questionnaire window. Those who chose to participate were asked to provide informed consent by checking a box at the bottom of the electronic consent form and start completing the questionnaire.

Four questions were used to measure the feasibility of administering the SLUMP questionnaire. The first question inquired about the length of the questionnaire. Students were asked to rate the length of the questionnaire on a scale of 0 to 10 (0 indicated too short and 10 indicated too long). The second question asked students to rate the clarity of the questions on a scale of 0 to 10 (0 indicated that the questions were unclear and 10 indicated that the questions were clear). The third question asked students to rate the ease of navigation of the questionnaire on a scale of 0 to 10 (0 indicated that the student could not navigate the questionnaire and 10 indicated that it was easy to navigate). Finally, students were asked to provide any additional comments regarding the questionnaire.

This study (REB File # 15-008) was reviewed by the University of Ontario Institute of Technology Research Ethics Board and was approved as of September 4th 2015. According to the ethics protocol, the completed questionnaires were de-identified by the UOIT IT department. A de-identified dataset was provided to the investigator by the IT department.

3 Analysis

We computed descriptive statistics for all variables included in the questionnaire to determine the presence of outliers. We used skip functions (for sections that were not applicable to participants) to improve the ease of completion of the questionnaire and reduce the participants’ burden. For example, if participants did not report using a laptop for employment purposes, they did not have to answer questions related to the specifics of using a laptop for employment purposes. Therefore, the total numbers of participants answering individual question may vary across the questionnaire. The analysis was conducted using SPSS version 23.0.

4 Results

The participation rate was 75% (33/44). Students took approximately 10–12 minutes to complete the questionnaire. Most participants were females (63.6%), the mean age was 19.79 years (CI 19.19, 20.38).

Participants took up to 12 minutes to complete the questionnaire. This was monitored by asking the class if they were done completing the questionnaire after 10 minutes and again after 12 minutes. Overall, participants indicated that the questionnaire was clear (mean = 8.6/10; CI = 8.0, 9.3) and easy to navigate (mean = 8.9/10 CI = 8.3, 9.4). The length of the questionnaire was perceived to be adequate (mean = 5.6/10 CI = 5.2, 6.1) (Table 1).

Table 1
Appraisal of Questionnaire, Length, Clarity and Ease of Navigation (n = 33)

Characteristics Mean Confidence Intervals Median Range

Length of Questionnaire (0–10)^a 5.64 5.19, 6.09 5 3–9

Clarity of Questions (0–10)^b 8.64 7.97, 9.30 10 3–10

Ease of Navigation (0–10)^b 8.85 8.25, 9.44 10 4–10

Characteristics	Mean	Confidence Intervals	Median	Range
Length of Questionnaire (0–10)^a	5.64	5.19, 6.09	5	3–9
Clarity of Questions (0–10)^b	8.64	7.97, 9.30	10	3–10
Ease of Navigation (0–10)^b	8.85	8.25, 9.44	10	4–10

Participants provided four comments about the SLUMP questionnaire: 1) “No need for improvement. It is just fine” (n = 1); 2) “the questions about taking breaks when using the computer need to be more clear” (n = 1); 3) “Allow half hour increments to be selected for certain time estimates (ex. time spent before taking break while on computer)” (n = 1); 4) “Use clearer subheadings when distinguishing between the questions for recreational use and educational use. Also add more options for number of hours to be more specific (eg. 0.5)” (n = 1).

4.1 Missing data

All participants completed the questionnaire without missing data.

4.2 Outliers

Only two outliers were present in the data. One student reported using a laptop at a desk for 24 hours per day and using a laptop for academic purposes for 24 hours per day. This suggests that overall the questionnaire allows for the collection of data within expected ranges.

4.3 Revisions to the SLUMP Questionnaire

We suggest multiple revisions to improve the questionnaire based on the comments provided by the participants. A revised version of the SLUMP questionnaire is available in Questionnaire C

Revision #1: The first comment we received was “No need for improvement. It is just fine.” As a result, we only made small changes to questions, which were identified as problems by other students.

Revision #2: The second comment we received suggested we “Allow half hour increments to be selected for certain time estimates (ex. time spent before taking break while on computer).” Consequently, we included an additional option for “every 0.5 hours” for questions regarding how often students take breaks.

Revision #3: The third and fourth comments we received were “Use clearer subheadings when distinguishing between the questions for recreational use and educational use. Also add more options for number of hours to be more specific (eg. 0.5)” and “The questions about taking breaks when using the computer need to be clearer.” To address these concerns, we capitalized all headers to ensure that the sections were distinguishable. The bigger letters will draw the reader’s attention to the words recreational, academic and employment use for each question.

5 Discussion

Our pilot study suggests that the SLUMP questionnaire is acceptable for in-class administration in university students. The questionnaire aims to measure laptop use in students and future administrations of the instrument may be completed during class time. In our sample, students completed the questionnaire within 12 minutes, which suggests that it is possible to administer it in-class with minimal disruption. Moreover, 75% of eligible students consented to participate which is similar to other questionnaires completed by university students which saw response rates ranging from 86% to 94% [4 , 25]. These students reported that the questionnaire was clear, easy to navigate and that its length was appropriate. Finally, our descriptive analysis suggests that the questionnaire captures the full range of possible response and that few outliers are present in the data. It is important for the questionnaire to capture a range of responses as Bubric & Hedge [26] identified that there are a number of different ways that laptops are used by post-secondary students.

Our pilot study suggests that the time to complete the SLUMP is similar to previously developed questionnaires. Specifically, the Laptop Use Scale requires 10–15 minutes to complete while the Boston University Computer and Health Survey takes between 15–20 to complete [4, 20].

All of the participants completed the SLUMP questionnaire without any missing data. The original administration of the SLUNPRQ had two questionnaires which were inadequately completed [21]. This suggests that the changes made to the SLUMP questionnaire reduce the amount of missing data provided by participants similar to other online questionnaires [27].

The test-retest reliability of items included in the SLUNPRQ ranged from kappa = 0.36–1.00. Grey [21] suggested that multiple revisions were necessary to improve the reliability of the SLUNPRQ. We made these modifications to the SLUNPRQ to improve the clarity of questions, quality of responses and ease of administration. We improve the clarity of the questions by providing examples of recreational laptop use, breaks and a brief description of the postures. Participants in the pilot study favorably rated the clarity of the SLUMP (mean = 8.6/10). Therefore, we feel our modifications improved the clarity of questions [28]. Similarly, adding a frame of reference for questions related to breaks may have helped students recall their laptop use more reliably [29].

The SLUMP questionnaire fills a void in the literature by evaluating ergonomic exposures to laptop use in university students. Although the Boston University Computer and Health Survey and Laptop Use Scale were designed to measure laptop use in students, neither questionnaire was developed to measure ergonomic exposures experienced by students when using a laptop [4, 20]. Additionally, the SLUMP questionnaire was developed to measure overall (including recreational and employment use) laptop use while the Laptop Use Scale only measures laptop use during class time [20].

Several studies suggest that laptop use is negatively associated with health [7 , 30]. Therefore, it is important to develop accessible methods to accurately measure exposures to laptop use in university students. Developing and pilot testing the SLUMP questionnaire is the first step in accurately measuring ergonomic exposures to laptop use in university students.

5.1 Limitations

Our pilot study has limitations. First, we did not collect data pertaining to why students did not participate in the study. Therefore, we are unable to comment on the reasons for non-participation and evaluate the presence of selection bias. Second, students suggested improving the clarity of some questions; therefore, it is possible that some of the questions in the pilot version of the questionnaire were associated with measurement error. Third, the SLUMP questionnaire was pilot-tested in one homogenous class. Therefore, the external validity of the results may be limited. Finally, our sample size of 33 students did not allow us to obtain preliminary data on exposures that are not relevant to all students (e.g., laptop use for employment purposes). Therefore we are unable to comment on whether the options we provided captured a range of responses.

5.2 Strengths

Our pilot study also has strengths. First, it provides a comprehensive (school, employment and recreation) measure of laptop use by students. Second, its design minimizes the burden on students. We included skip functions to reduce the completion time by students. Third the questionnaire is clear and easy to administer in a classroom.

6 Conclusions

The results from our pilot study suggests that it is feasible to administer the Student Laptop Use and Musculoskeletal Posture questionnaire in-class to university students. Future research should evaluate the test-retest reliability of the revised SLUMP questionnaire. The validity of this questionnaire should also be assessed before the questionnaire is used. Reliability and validity can contribute to adding objectivity to a research design, so we highly recommend further evaluations of this questionnaire.

Conflict of interest

None to report.

Footnotes

Appendix

References

Caruso

, Salaway

. The ECAR study of undergraduate students and information technology.The Educause Center for Applied Research 2007.

Dockrell

, Bennett

, Culleton-Quinn

. Computer use and musculoskeletal symptoms among undergraduate university students. Comput Educ 2015;85, 102–9.

Kandri

, Bonois

, Floros

, Zafiropoulou

. Alexithymia components in excessive internet users: A multi-factorial analysis. Psychiatry Res 2014;220(1):348–55.

Obembe

, Johnson

, Tanimowo

, Onigbinde

, Emechete

. Musculoskeletal pain among undergraduate laptop users in a Nigerian University. J Back Musculoskelet Rehabil 2013;26(4):389–95.

Reddy

, Low

, Lim

, Low

, Mardina

, Nursaleha

. Computer vision syndrome: A study of knowledge and practices in university students. Nepal Ophthalmic Society 2013;5(2):161–8.

Rempel

, Keir

, Bach

. Effect of wrist posture on carpal tunnel pressure while typing. J Orthop Res 2008;26(9):1269–73.

Kanchanomai

, Janwantanakul

, Pensri

, Jiamjarasrangsi

. Prevalence of and factors associated with musculoskeletal symptoms in the spine attributed to computer use in undergraduate students. Work 2012;43(4):497–506.

Leccese

, Salvadori

, Rocca

. Visual discomfort among university students who use CAD workstations. Work 2016;55(1):171–80.

Gold

, Driban

, Yingling

, Komaroff

. Characterization of posture and comfort in laptop users in non-desk settings. Appl Ergon 2012;43(2):392–9.

10.

Rudolf

, Griffiths

. Evaluating the ergonomics of a student learning environment. Work 2009;34(4):475–80.

11.

Szeto

, Lee

. An ergonomic evaluation comparing desktop, notebook, and subnotebook computers. Arch Phys Med Rehabil 2002;83(4):527–32.

12.

Werth

, Babski-Reeves

. Effects of portable computing devices on posture, muscle activation levels and efficiency. Appl Ergon 2014;45(6):1603–9.

13.

Taib

, Bahn

, Yun

. The effect of psychosocial stress on muscle activity during computer work: Comparative study between desktop computer and mobile computing products. Work 2016;54(3):543–55.

14.

Baker

, Cook

, Redfern

. Rater reliability and concurrent validity of the Keyboard Personal Computer Style instrument (K-PeCS). Appl Ergon 2009;40(1):136–44.

15.

Siu

DCH

, Tse

, Yu

ITS

, Griffiths

. Computer products usage and prevalence of computer related musculoskeletal discomfort among adolescents. Work 2009;34(4):449–54.

16.

Speklé

, Hoozemans

, van der Beek

, Blatter

, van Dieën

. The predictive validity of the RSI QuickScan questionnaire with respect to arm, shoulder and neck symptoms in computer workers. Ergonomics 2012;55(12):1559–70.

17.

Menéndez

, Amick

III BC

, Jenkins

, Janowitz

, Rempel

, Robertson

M et al

. A multi-method study evaluating computing-related risk factors among college students. Work 2007;28(4):287–97.

18.

Perreault

, Brisson

, Dionne

, Montreuil

, Punnett

. Agreement between a self-administered questionnaire on musculoskeletal disorders of the neck-shoulder region and a physical examination. BMC Musculoskelet Disord 2008;9(1):34.

19.

Jacobs

, Foley

, Punnett

, Hall

, Gore

, Brownson

, . University students’ notebook computer use: Lessons learned using e-diaries to report musculoskeletal discomfort. Ergonomics 2011;54(2):206–19.

20.

Kay

, Lauricella

. Assessing laptop use in higher education: The Laptop Use Scale. JCHE 2015;27(3):1–27.

21.

Gray

. Development of Instruments to Assess Physiological and Physical Neck Pain Risk Factors [Unpublished master’s thesis]. University of Ontario Institute of Technology, Oshawa, Ontario 2011.

22.

Hohwü

, Lyshol

, Gissler

, Jonsson

, Petzold , M , Obel

. Web-based versus traditional paper questionnaires: A mixed-mode survey with a Nordic perspective. J Med Internet Res. 2013;15(8):173.

23.

Horevoorts

, Vissers

, Mols

, Thong

, van de Poll-Franse

. Response rates for patient-reported outcomes using web-based versus paper questionnaires: Comparison of two invitational methods in older colorectal cancer patients. J Med Internet Res. 2015;17(5).

24.

Sebri

, Bartier

, Pelaccia

. How Do Nursing Students Use Digital Tools during Lectures? PloS one. e 2016;11(11):0165714.

25.

Ryan

, Marley

, Still

, Lyons

, Hood

. Use of mental-health services by Australian medical students: A cross-sectional survey. Australasian Psychiatry. Aug 2017;1039856217715990.

26.

Bubric

, Hedge

. Differential patterns of laptop use and associated musculoskeletal discomfort in male and female college students. Work 2016;55(3):1–9.

27.

Rayhan

, Zheng

, Uddin

, Timbol

, Adewuyi

, Baraniuk

. Administer and collect medical questionnaires with Google documents: A simple, safe, and free system. Applied Medical Informatics 2013;33(3):12.

28.

Streiner

, Norman

, Cairney

. Health measurement scales: A practical guide to their development and use Oxford University Press, USA, 2015.

29.

Stone

, Turkkan

, Bachrach

, Jobe

, Kurtzman

, Cain

. The science of self-report: Implications for research and practice. Lawrence Erlbaum Associates. New Jersey. 2000.

30.

Chiou

, Chou

, Chen

. Notebook computer use with different monitor tilt angle: Effects on posture, muscle activity and discomfort of neck pain users. Work 2012;41(Supplement 1):2591–5.