Share this article on:

The Second Version of the L. V. Prasad-Functional Vision Questionnaire

Gothwal, Vijaya K.*; Sumalini, Rebecca; Bharani, Seelam; Reddy, Shailaja P.; Bagga, Deepak K.§

doi: 10.1097/OPX.0b013e31826ca291
Original Articles

Purpose. The L. V. Prasad-Functional Vision Questionnaire (LVP-FVQ) was developed using Rasch analysis to assess self-reported difficulties in performing daily tasks in school children with visual impairment (VI) in India. However, the LVP-FVQ has psychometric problems of inadequate measurement precision and lack of detailed assessment of dimensionality. Furthermore, items pertaining to use of technology are lacking. The aim of this study was to present the development and validation of the second version of LVP-FVQ (LVP-FVQ II).

Methods. Development of LVP-FVQ II involved extracting items from other similar questionnaires (albeit developed for Western populations) and focus group discussions of children with VI and their parents that resulted in a 32-item pilot questionnaire. Overall, six items from the LVP-FVQ were retained. The questionnaire underwent pilot testing in 25 such children, following which a 27-item LVP-FVQ II emerged, and this was administered to 150 children with VI. Response to each item was rated on a three-category scale. Rasch analysis was used to validate the LVP-FVQ II.

Results. Rating scale was used by participants as was intended to. Four mobility-related items required deletion, as these did not contribute toward measurement of a single construct, indicating a secondary dimension. Deletion of the four items resulted in the 23-item unidimensional LVP-FVQ II, with good measurement precision, effective targeting of item difficulty to participant ability, and lack of notable differential item functioning. The LVP-FVQ II has high reliability, indicating that it is effectively able to discriminate between visual disability of school children in India, and is valid across age, gender, duration of VI, and location of residence.

Conclusions. Given the superior measurement properties and the interval-level scores, the LVP-FVQ II appears to offer advantages over LVP-FVQ in assessment of difficulties in performing daily tasks in this population. It can be adapted for use in other developing countries.



BSOpt, MPhil(Opt)


Meera and L. B. Deshpande Centre for Sight Enhancement, Vision Rehabilitation Centres, L. V. Prasad Eye Institute, Hyderabad, India.

The L. V. Prasad-Functional Vision Questionnaire (LVP-FVQ) was developed by us in 2003 using Rasch analysis to assess the self-reported difficulties in performing daily tasks in school-going children (8 to 17 years) with visual impairment (VI) in developing countries such as India.1 Since its development, the LVP-FVQ has been used for research in India as well as in the Middle East. For example, the LVP-FVQ was used in a population-based study primarily designed to determine appropriate service delivery models for eye care services targeting children in rural communities in South India.2 However, a shorter version of the LVP-FVQ (12 items) was used that was demonstrated to possess superior psychometric properties using Rasch analysis as compared with the original version. Recently, the 20-item LVP-FVQ was translated into Arabic for use in a hospital-based study of Egyptian children with VI.3 These investigators also demonstrated the LVP-FVQ to be useful to assess the functional difficulties of children with VI, but they did not report its psychometric properties in detail.

Our previous Rasch analysis of the LVP-FVQ indicated the need for some modifications. First, a reduction in number of categories that was further confirmed in the Egyptian study. Second, the measurement precision was lower than acceptable, suggesting that the questionnaire was unable to reliably distinguish among participants’ visual abilities. Third, there was an item (“applying paste on a brush”) that displayed large misfit, perhaps indicating at the possibility of an underlying additional dimension being measured by this item (i.e., there may be a lack of unidimensionality). In addition, a detailed assessment of the dimensionality of the LVP-FVQ was not performed in the original analysis. Fourth, a need for additional mobility-related items was recognized. Furthermore, given that the LVP-FVQ was developed about a decade ago, items related to use of technology are lacking. That is, the range of activities in the LVP-FVQ is limited in that they do not encompass some of those activities that are performed by current generation of children, for example, using mobile phones, using computers, and playing video games. Given the boom in the information communication technology sector in India over the past few years, there has been an increasing interest and growing demand for the use of technology in the society, and children are being continuously exposed to electronic gadgets both at school and at home.4 Thus, there is a shift in the visual requirements of children under these prevailing circumstances so as to be able to accomplish tasks that involve use of technology. This suggests that a questionnaire developed to assess their visual abilities should include items related to such activities. However, addition of items cannot be an ad hoc process and should not be taken lightly. Rather items can be added, but the modified version of the questionnaire has to be re-validated, simulating the development of a new questionnaire. The revised version should not only be better on a theoretical basis, but also have superior psychometric properties than the original version.

Recently, two questionnaires, namely, the Impact of Vision Impairment for Children (IVI_C)5 and the Cardiff Visual Ability Questionnaire for Children (CVAQC),6 have been published for use in children with VI in the developed countries. By comparison, the LVP-FVQ remains till date as the single most developed questionnaire for use in Indian children with VI. The lack of availability of other questionnaires, coupled with the long overdue need to incorporate the modifications in the LVP-FVQ from our previous analyses, convinced us that the concept has to be further developed for use in this part of the world. Furthermore, the lack of availability of simple scoring conversions of raw ordinal data to interval-level scores may be preventing its widespread use by researchers and clinicians alike in India. Therefore, an improved version of the questionnaire with a simpler scoring method may further enhance its use in research and outcome assessments, for example, in clinical trials involving interventions for children with VI. Thus, the primary aim of the present study was to describe the development and comprehensive validation of the LVP-FVQ II (including dimensionality) using Rasch analysis in children with VI in India. Our secondary aim was to provide ready-to-use spreadsheets that convert raw scores to Rasch-scaled scores for the LVP-FVQ II for the benefit of those who are unfamiliar with Rasch analysis but wish to use its scoring advantages.

Back to Top | Article Outline



Participants were children who were referred to Vision Rehabilitation Centres at the L. V. Prasad Eye Institute (LVPEI), Hyderabad, India, for management of low vision. Details of the low vision rehabilitation program at Vision Rehabilitation Centres, LVPEI, have been published earlier in this journal.7 Included participants were those who had self-reported difficulty with performing day-to-day tasks, enrolled in regular schools (grades 3 to 10), could perform standard clinical vision tests, and were able to respond to the items (questions) on the LVP-FVQ. We included children with visual acuity of 20/20 (n = 36) because they had accompanying visual field loss in the better eye (<20°). Children with other disabilities (such as sensory, physical, and/or intellectual), previous users of low vision devices, and those unable to understand English, Hindi, or Telugu were excluded from the study. The study received approval from the Ethics committee of the LVPEI, and the research adhered to the tenets of the Declaration of Helsinki. Informed consent was obtained from the participants and their parents or caregivers after a detailed explanation of the study.

Back to Top | Article Outline


L. V. Prasad-Functional Vision Questionnaire II

The development of the LVP-FVQ II took place in two main phases: (i) questionnaire development, and (ii) psychometric analyses. Each of these phases is described in the text.

Back to Top | Article Outline
Phase I: Questionnaire development

In the first step, we reviewed all the existing questionnaires for use in children with VI regarding the suitability of content to determine whether items could be borrowed for use in the development of the LVP-FVQ II. In the second step, we conducted four focus group discussions (FGDs) to identify activities of children specific to Indian context that they or their parents considered to be affected owing to their visual condition. We conducted separate FGDs for parents and for their children with VI (two each) because the advantage of such group composition would ensure retention of the group dynamics and elimination of any confounding effect of parent/proxy responses and bias toward information solely (or majority) from parents. Each FGD lasted 1 to 2 h; the group included six to eight children with VI aged 8 to 16 years, and their visual acuities spanned over a wide range from mild to severe VI. The FGDs with parents comprised 10 participants in each group. The FGDs were conducted in a room (away from the clinic), where the participants were seated around a square table. An experienced low vision specialist (VKG) with previous experience in conducting the FGDs moderated the sessions (assisted by the second author) that were audio-taped to facilitate content analysis. An abridged transcription of the audio tape recording was also carried out by the moderator. The exact expressions used by the participants were, however, conserved in the transcripts.

The FGDs were followed by eight semi-structured interviews done separately for children with VI and their parents who could not attend the FGDs.

The data were analyzed qualitatively based on the grounded theory approach.8 Briefly, the audiotapes were initially transcribed verbatim independently by both the moderator and second author. After this, individual meaningful statements were constructed (albeit with paraphrasing) from these transcripts to identify the list of themes/codes from the comments of the participants. Statements with common themes were grouped to form major themes—activities of daily living, academic and leisure activities. Subthemes were subsequently identified and classified under the relevant major themes. This was done for each type of focus group. The themes from the semi-structured interviews were not, however, distinct from those in the FGDs of parents and their children. When the analyses were completed, both the moderator (expert) and the second author met up to compare the agreement and discrepancies between the two analyses for each type of focus group. Areas of discrepancies were highlighted and resolved through discussion. In the event that both were unable to reach a consensus, a neutral third-party opinion was sought from the other authors who were asked to join in the discussion. All discrepancies were resolved before finalizing the items for inclusion in the questionnaire. The processes from steps 1 and 2 resulted in a 32-item pilot questionnaire. Of these 32 items, 6 belonged to the LVP-FVQ, 2 to the IVI_C, and 4 to the CVAQC. Although these items were borrowed from existing questionnaires including the original LVP-FVQ, minor modifications such as rephrasing (for e.g., “locating dropped objects like pen, pencil, rubber within the classroom” was changed to “locating dropped objects” to make it non-specific) were required for inclusion in the LVP-FVQ II. The questionnaire was initially developed in Indian English and was later translated and back-translated into two local languages, i.e., Telugu and Hindi, using standard accepted procedures.

The 32-item pilot questionnaire was tested in a representative sample of 25 children with VI studying in grades 3 to 10. Six items were discarded, as they had a skewed distribution, with 80% of the participants reporting “no difficulty” with these activities (Table 1). Therefore, the final version of LVP-FVQ II consisted of 27 items, i.e., 26 plus a global rating item (Table 2). We changed the response options to a three-category scale (1, no difficulty; 2, some difficulty; and 3, a lot of difficulty). Like the original LVP-FVQ, an option of “not applicable” was used for all the items (except the global rating item), and the global rating item had different response options as compared with the bulk of the items. The response options of the global rating item were as follows: 1, as well as your friend’s; 2, a little bit worse than your friend’s; 3, much worse than your friend’s. Unlike the previous Rasch analysis of the LVP-FVQ, we included the global rating of vision in the calculation of the total LVP-FVQ II score.





Back to Top | Article Outline
Phase II: Psychometric analyses
LVP-FVQ II Study Sample

One hundred seventy-three children with VI were recruited for this phase of the study. The LVP-FVQ II was administered to each participant in a face-to-face interview. Twenty-three participants were administered the questionnaire twice over an average duration of 2 months for an assessment of temporal stability. The retest participants were those who had not adhered to the low vision management (such as purchase/use of low vision devices or any other suggestion that was provided) so as to maintain similar testing conditions across questionnaire administration.

Back to Top | Article Outline

Data Collection: Demographic Data and Vision Measures

Sociodemographic data including age, gender, and cause of VI were ascertained from the medical records. Habitual monocular and binocular visual acuity was measured on all participants using a high-contrast letter chart based on logMAR principles. Given that binocular acuity is considered to be representative of real world ability, we used it in all our analyses.9,10

Back to Top | Article Outline

Statistical Analysis

We used SPSS software (ver. 16.0) to conduct the descriptive analysis. P < 0.05 was considered as statistically significant. Criterion validity was assessed by determining the relationship between the Rasch-scaled LVP-FVQ II person scores (Rasch analysis described in detail later in the article) and presenting high-contrast binocular visual acuity. We used intraclass correlation coefficient (ICC) to determine the test–retest reliability (temporal stability) in a two-way mixed-effects model.11,12 ICC values >0.75 were interpreted as good reliability, whereas values <0.75 indicated poor to moderate reliability.13

Back to Top | Article Outline

Rasch Analysis

Rasch analysis14 was conducted using the Andrich rating scale model15 with Winsteps software (version 3.68).16 As mentioned in the Methods section, the items used two different rating scales requiring the use of the two Andrich rating scale model (one for items 1 to 26 and another for item 27). This approach has been described earlier.17,18 Rasch analysis is an iterative procedure that estimates interval measurement from ordinal data, and the unit is logits (log-odd units).18–20 For the present study, a negative person logit indicates that the participant’s ability is higher than the mean visual ability required for all the items. A negative item logit indicates that the required visual ability for that item is higher than the mean required visual ability of all the items. The Rasch procedures have been described in detail previously in this journal.7,20 In brief, four fundamental indicators were used to evaluate questionnaire quality.21 These included (i) fit, or the extent that items in the questionnaire measured a single construct (i.e., unidimensionality) using fit statistics (infit mean square, MnSq; acceptable range: 0.7 to 1.3) and principal components analysis (PCA) of residuals,20,22 (ii) item difficulty, or the hierarchical ordering of items from “more to less difficult,” and targeting, or the extent to which the item difficulty match with participant abilities visualized using the person-item map (mistargeting; >1.0 logit23), (iii) measurement precision using person separation reliability (PSR), with minimum acceptable PSR of 0.80 that indicates the ability of the questionnaire to distinguish among at least three strata of visual disability in the participants,22,24 and (iv) differential item functioning (DIF; notable if >1.0 logit22) for age, gender, duration of VI, and location of residence.

Back to Top | Article Outline


Participants’ Sociodemographic Characteristics

We administered the 27-item questionnaire to 173 children. The mean age of the participants was 12.2 years (SD = 2.5; range = 8 to 16), and the majority (n = 103, 60%) was male. The sociodemographic characteristics of the participants are provided in Table 3.



Back to Top | Article Outline

Psychometric Validation of the LVP-FVQ II

Rating Scale Analysis

The questionnaire demonstrated a well-functioning rating scale, indicating that the participants used the categories as were intended to.

Back to Top | Article Outline

Measurement Precision and Targeting

The PSR was 0.83, implying that the questionnaire could differentiate among at least three groups of participant’s visual abilities (Table 4). The mean participant ability was −0.88 logits, indicating the item difficulties were matched well to the participant visual abilities.



Back to Top | Article Outline


All the 27 items fit the Rasch model well (infit MnSq, 0.72 to 1.28). However, PCA of residuals showed that the variance explained by the measures was 44.1% by empirical calculation and 41.2% by the model. These values are lower than the minimum acceptable value of 60%. Additionally, the unexplained variance by the first contrast was 2.4 eigenvalue units, but no further contrasts exceeded 2.0 eigenvalue units (Table 4). There were four items in the first contrast with loading >0.4, and all these items represented mobility (1, “walking without bumping into objects/people”; 2, “getting around alone in places you know at nighttime”; 3, “moving around safely in places you don’t know in the day time”; 4, “moving around in places you don’t know at nighttime”). The results of PCA of residuals indicated lack of unidimensionality that was being caused by the mobility-related items. Establishing unidimensionality is essential for validity of the questionnaire, so the way forward was to delete the mobility-related items. After item deletion, Rasch analysis was re-run on the remaining 23 items. The PCA of residuals now showed that the unexplained variance by the first contrast was 1.8 eigenvalue units (Table 4). All the 23 items in the LVP-FVQ II fit the Rasch model, indicating that it was a unidimensional measure (Table 5). The mean participant ability was −0.90 logits, and the PSR was 0.83. Given the adequate PSR that is an improvement from the original LVP-FVQ, we were interested in determining whether the inclusion of the global rating item had contributed to this enhanced reliability. However, we found the PSR to be equally good (0.81) even without this item.



The person-item map in Fig. 1 displays participants’ scores on logit scale (left) and the relative difficulty levels of each of the items (right). Participants with the higher visual ability and most difficult items are located toward the bottom of the map. The three most difficult items were “your vision compared to your normally sighted friend,” “copying the small letters from the board,” and “threading a needle.” The three least difficult items were “watching movie at the theater,” “identifying dirt, stains on your clothes,” and “selecting a song using iPod.” The deleted items were assessed for fit to the Rasch model to determine whether they could form a separate reliable and valid scale. However, these items lacked adequate measurement precision (PSR = 0.29).



Back to Top | Article Outline

Differential Item Functioning

None of the items showed notable DIF by the variables assessed.

Back to Top | Article Outline

Criterion Validity

There was a moderate significant linear relationship (r = 0.51, p < 0.0001) between the presenting binocular high-contrast visual acuity and the Rasch-scaled LVP-FVQ-II score that accounted for 25.2% of the variance in the participant’s visual abilities.

Back to Top | Article Outline

Test–Retest Reliability

Fig. 2 is a scatterplot of person measures estimated from responses to the initial administration of the LVP-FVQ II by 23 participants vs. person measures estimated from person responses to the re-administration of the same questionnaire after an average duration of 2 months. There was good test–retest concordance, with an ICC of 0.89 for persons (95% CI, 0.76 to 0.95). Fig. 3 is a scatterplot for item measures. The ICC for item measures was 0.95 (95% CI, 0.88 to 0.98), indicating good test–retest concordance.





Back to Top | Article Outline

Raw Score to Rasch Measure Conversion

To assist users who wish to use the benefits of Rasch analysis in the scoring of the LVP-FVQ II, but are unfamiliar with the procedure, we have developed ready-to-use spreadsheets for conversion of raw scores to Rasch-scaled scores for the questionnaire. These sheets can be obtained by contacting the corresponding author or can be downloaded from the journal’s Web site. However, we caution that these conversions can be applied only if the population on which it is being tested is similar to that of the present study.

Back to Top | Article Outline


Our results have demonstrated that the LVP-FVQ II has excellent psychometric properties: unidimensionality, good reliability, adequate targeting, and a well-functioning rating scale. Furthermore, the LVP-FVQ II demonstrated good test–retest reliability over time in our study population.

A balance has to be maintained between conservatism and innovation in the development of questionnaires. If a questionnaire is revised frequently, comparisons between studies are hampered, as the items will not be identical. On the other hand, if problems have been identified, revisions to improve the validity and reliability seem logical. The development of the LVP-FVQ II was based on sound theoretical considerations and our own experience in the clinic regarding children with VI.

Although we used Rasch analysis in our earlier and present study of the LVP-FVQ, our previous analysis lacked an assessment of dimensionality and DIF of the questionnaire. Rasch analysis examines the unidimensionality of the questionnaire, i.e., the extent to which all items in the questionnaire measure the same underlying construct and how well each item measures or “fits” the construct.21 The 27-item LVP-FVQ II lacked unidimensionality as was evidenced by PCA of residuals, thereby violating the fundamental requirement of the Rasch measurement model. Furthermore, the PCA of residuals indicated that the questionnaire was measuring more than construct, i.e., secondary dimension—mobility. Other investigators have also found items related to mobility (and driving) that do not fit the core set of items.17,22 Unidimensionality is critical to a questionnaire because in its absence, the user is unsure of the construct under measurement, so it is mandatory that it is re-established for a optimally functioning questionnaire. Consequently, four mobility items were deleted to restore unidimensionality of the LVP-FVQ II. Unlike the earlier Rasch analysis of the LVP-FVQ with a relatively smaller sample (n = 78), the larger sample size in the present study (n = 173) helped to estimate stable and more precise item calibrations, enabling the determination of a satisfactory solution. Thus, the LVP-FVQ II total or summary score is valid with interval-level properties that now legitimizes the use of parametric statistics for calculation of change scores,26 for example, comparison of pre and post low vision rehabilitation intervention data.

The other desirable features of an optimally functioning questionnaire include an ability to discriminate among as many groups of participant ability as possible, and higher the number of strata, the better is the discriminative ability of the questionnaire.24,26 The PSR of the LVP-FVQ II is excellent, indicating that it is reliably able to differentiate among different strata of participant’s visual abilities. The test–retest data of the LVP-FVQ II have demonstrated that it has excellent levels of stability over a 2-month period.

Similar to the LVP-FVQ, its revised version—23-item LVP-FVQ II—also demonstrated adequate targeting (−0.90 logits) of item difficulty to participant ability. One of the reasons for this adequate targeting could be related to the generation of item content using FGDs. FGDs help keep the item content relevant to the population under consideration and have been used by other investigators as well.5,6,27 Given that FGDs involved a sample of school children with VI and their parents, the finding of adequate targeting of the LVP-FVQ II should not be surprising. Nonetheless, it can be seen from the person-item map in Fig. 1 that some items share the same item location, perhaps indicating redundancy, but item deletion may affect measurement precision. Given the nature of item generation (using FGDs), we decided to retain all the items for the present. Although questionnaires can be periodically revised to suit the changing visual demands of the population, the best way forward to circumvent the problem of the need for constant revisions would be the development of item banks.28–30 Item banks can be administered using computer-adaptive testing, whereby the answer limited the questions based on the response to the preceding question.31 Although item banks have been developed and used successfully in other fields of health care,32,33 such a concept is in its nascent stage in ophthalmology. However, attempts are being made in this direction, and item banks for use with adult population have been constructed.30,34 Similar item banks for children with VI require to be developed.

In our sample, the three most difficult items reported by children with VI included “your vision compared to your normally sighted friend,” “copying the small letters from the board,” and “threading a needle.” Of these, the item “threading a needle” was common to the previous analysis. Both the tasks—“copying small letters from the board” and “threading a needle”—are those that require high resolution, so this item hierarchy was expected. In our earlier analysis, we did not include the global rating item in the calculation of the total questionnaire score. However, this item was rated as the most difficult in that children found it difficult to choose lowest response category (better score). It is not unusual to include the global rating item along with other items in the calculation of the total or summary score. Both the original 39- and 25-item National Eye Institute—Visual Functioning Questionnaire and their recently proposed corresponding Rasch versions, i.e., the 12-item and 8-item long form visual functioning scales include the global rating item in the calculation of the total score.18,27 In our previous analysis, we had shown this item to possess high discrimination ability, and its inclusion in the precision study increased the measurement precision (albeit marginally; 0.81 without this item, 0.83 with this item). Highly discriminating items are essential in a questionnaire so as to improve the measurement precision. Taken together, we believe that it is appropriate to include this item in the calculation of the total LVP-FVQ II score.

The response scale of the LVP-FVQ II consisting of three options functioned well in that the participants made use of all the categories. Smaller numbers of categories (three) are potentially less confusing for the respondents, as shown for questionnaires developed for adults.7,20 Questionnaires developed for children in the West have used >3 response options, and investigators have had to perform category re-organization to improve the performance of the rating scale.5,6

We have developed a robust questionnaire, the LVP-FVQ II, which meets the quality assessment criteria that have been proposed for appropriate development, refinement, and validation of a questionnaire.24 The 23-item LVP-FVQ II has some major advantages in terms of possessing superior psychometric properties including unidimensionality and reliability than the native LVP-FVQ. We recommend replacing the LVP-FVQ with the second version, i.e., LVP-FVQ II. Given that we have provided conversion sheets, it is now simpler to score; we encourage investigators to use the LVP-FVQ II. Further studies assessing the responsiveness of the LVP-FVQ II in school children with VI to low vision interventions are required, and one such study is currently underway at our center.

Vijaya K. Gothwal

Meera and L. B. Deshpande Centre for Sight Enhancement

Vision Rehabilitation Centres, L. V. Prasad Eye Institute

Kallam Anji Reddy Campus, L. V. Prasad Marg

Banjara Hills, Hyderabad 500034 Andhra Pradesh India


Back to Top | Article Outline


This study was supported by the Hyderabad Eye Research Foundation, Hyderabad, India.

The authors have no personal financial interest in the development, production, or sale of any device discussed herein.

Received April 14, 2012; accepted July 10, 2012.

Back to Top | Article Outline


1. Gothwal VK, Lovie-Kitchin JE, Nutheti R. The development of the LV Prasad-Functional Vision Questionnaire: a measure of functional vision performance of visually impaired children. Invest Ophthalmol Vis Sci 2003; 44: 4131–9.
2. Nirmalan PK, John RK, Gothwal VK, Baskaran S, Vijayalakshmi P, Rahmathullah L. The impact of visual impairment on functional vision of children in rural South India: the Kariapatti Pediatric Eye Evaluation Project. Invest Ophthalmol Vis Sci 2004; 45: 3442–5.
3. Atasavun Uysal S, Turan A. Relationship between visual perception and functional visual questionnaire in children with low vision. Presented at the 7th European Conference of International Council for Education of People with Visual Impairment; 2009; Dublin, Ireland.
4. Pawar JB, Badiger CA, Nagnur S. Effects of computers on social life of children and adolescent. Karnataka J Agric Sci 2011; 24: 375–7.
5. Cochrane GM, Marella M, Keeffe JE, Lamoureux EL. The Impact of Vision Impairment for Children (IVI_C): validation of a vision-specific pediatric quality-of-life questionnaire using Rasch analysis. Invest Ophthalmol Vis Sci 2011; 52: 1632–40.
6. Khadka J, Ryan B, Margrain TH, Court H, Woodhouse JM. Development of the 25-item Cardiff Visual Ability Questionnaire for Children (CVAQC). Br J Ophthalmol 2010; 94: 730–5.
7. Marella M, Gothwal VK, Pesudovs K, Lamoureux E. Validation of the visual disability questionnaire (VDQ) in India. Optom Vis Sci 2009; 86: 826–35.
8. Strauss A, Corbin J. Grounded theory methodology: an overview. In: Denzin NK, Lincoln YS, ed. Handbook of Qualitative Research. Thousand Oaks, CA: Sage Publications; 1994: 273–85.
9. Rubin GS, Bandeen-Roche K, Huang GH, Munoz B, Schein OD, Fried LP, West SK. The association of multiple visual impairments with self-reported visual disability: SEE project. Invest Ophthalmol Vis Sci 2001; 42: 64–72.
10. Elliott DB, Hurst MA, Weatherill J. Comparing clinical tests of visual function in cataract with the patient’s perceived visual disability. Eye 1990; 4: 712–17.
11. Deyo RA, Diehr P, Patrick DL. Reproducibility and responsiveness of health status measures. Statistics and strategies for evaluation. Control Clin Trials 1991; 12: 142S–58S.
12. Muller R, Buttner P. A critical discussion of intraclass correlation coefficients. Stat Med 1994; 13: 2465–76.
13. Portnoy LG, Watkins MP. Foundations of Clinical Research: Applications to Practice, 2nd ed. Upper Saddle River, NJ: Prentice Hall Health; 2000.
14. Rasch G. Probabilistic Models for Some Intelligence and Attainment Tests (Expanded ed). Chicago, IL: University of Chicago Press; 1960.
15. Andrich DA. A rating scale formulation for ordered response categories. Psychometrika 1978; 43: 561–73.
16. Linacre JM. WINSTEPS Rasch measurement computer program. Chicago, IL:; 2009.
17. Gothwal VK, Wright TA, Lamoureux EL, Pesudovs K. Rasch analysis of visual function and quality of life questionnaires. Optom Vis Sci 2009; 86: 1160–8.
18. Pesudovs K, Gothwal VK, Wright T, Lamoureux EL. Remediating serious flaws in the National Eye Institute Visual Function Questionnaire. J Cataract Refract Surg 2010; 36: 718–32.
19. Massof RW, Fletcher DC. Evaluation of the NEI visual functioning questionnaire as an interval measure of visual ability in low vision. Vision Res 2001; 41: 397–413.
20. Gothwal VK, Wright TA, Lamoureux EL, Pesudovs K. Rasch analysis of the quality of life and vision function questionnaire. Optom Vis Sci 2009; 86: 836–44.
21. Bond TG, Fox CM. Applying the Rasch Model: Fundamental Measurement in the Human Sciences. Mahwah, NJ: Lawrence Erlbaum Associates; 2001.
22. Gothwal VK, Wright TA, Lamoureux EL, Pesudovs K. Activities of Daily Vision Scale: what do the subscales measure? Invest Ophthalmol Vis Sci 2010; 51: 694–700.
23. Mallinson T, Stelmack J, Velozo C. A comparison of the separation ratio and coefficient alpha in the creation of minimum item sets. Med Care 2004; 42: I17–24.
24. Pesudovs K, Burr JM, Harley C, Elliott DB. The development, assessment, and selection of questionnaires. Optom Vis Sci 2007; 84: 663–74.
25. Tennant A, McKenna SP, Hagell P. Application of Rasch analysis in the development and application of quality of life instruments. Value Health 2004; 7 (Suppl 1): S22–6.
    26. Mallinson T. Why measurement matters for measuring patient vision outcomes. Optom Vis Sci 2007; 84: 675–82.
    27. Mangione CM, Berry S, Spritzer K, Janz NK, Klein R, Owsley C, Lee PP. Identifying the content area for the 51-item National Eye Institute Visual Function Questionnaire: results from focus groups with visually impaired persons. Arch Ophthalmol 1998; 116: 227–33.
    28. Hahn EA, Cella D, Bode RK, Gershon R, Lai JS. Item banks and their potential applications to health status assessment in diverse populations. Med Care 2006; 44: S189–97.
    29. Bode RK, Lai JS, Cella D, Heinemann AW. Issues in the development of an item bank. Arch Phys Med Rehabil 2003; 84: S52–60.
    30. Pesudovs K. Item banking: a generational change in patient-reported outcome measurement. Optom Vis Sci 2010; 87: 285–93.
    31. Cook KF, O’Malley KJ, Roddey TS. Dynamic assessment of health outcomes: time to let the CAT out of the bag? Health Serv Res 2005; 40: 1694–711.
    32. Holman R, Lindeboom R, Vermeulen M, de Haan RJ. The AMC Linear Disability Score project in a population requiring residential care: psychometric properties. Health Qual Life Outcomes 2004; 2: 42.
    33. Lai JS, Cella D, Chang CH, Bode RK, Heinemann AW. Item banking to improve, shorten and computerize self-reported fatigue: an illustration of steps to create a core item bank from the FACIT-Fatigue Scale. Qual Life Res 2003; 12: 485–501.
    34. Fenwick EK, Pesudovs K, Khadka J, Dirani M, Rees G, Wong TY, Lamoureux EL. The impact of diabetic retinopathy on quality of life: qualitative findings from an item bank development project. Qual Life Res, in press.

    Rasch analysis; India; questionnaire; L. V. Prasad – Functional Vision Questionnaire; development

    © 2012 American Academy of Optometry