Free
Low Vision  |   July 2012
National Eye Institute Visual Function Questionnaire or Indian Vision Function Questionnaire for Visually Impaired: A Conundrum
Author Notes
  • From the Meera and L B Deshpande Centre for Sight Enhancement, Vision Rehabilitation Centres, L V Prasad Eye Institute, Hyderabad, India. 
  • Corresponding author: Vijaya K. Gothwal, Meera and L B Deshpande Centre for Sight Enhancement, Vision Rehabilitation Centres, L V Prasad Eye Institute, Hyderabad - 500034, Andhra Pradesh, India; vijayagothwal@gmail.com
Investigative Ophthalmology & Visual Science July 2012, Vol.53, 4730-4738. doi:10.1167/iovs.11-8776
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Vijaya K. Gothwal, Shailaja P. Reddy, Rebecca Sumalini, Seelam Bharani, Deepak K. Bagga; National Eye Institute Visual Function Questionnaire or Indian Vision Function Questionnaire for Visually Impaired: A Conundrum. Invest. Ophthalmol. Vis. Sci. 2012;53(8):4730-4738. doi: 10.1167/iovs.11-8776.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose.: Both the long form visual functioning scale (LFVFS39) and visual functioning scale (VFS) are measures of visual functioning (VF) that represent the Rasch-scaled versions of the NEI-VFQ39 and the Indian vision function questionnaire (IND-VFQ), respectively. The objectives of this study were to investigate if the 15-item LFVFS39 and 13-item VFS of the IND-VFQ meet the assumptions of the Rasch model and measure the same construct, VF, in an Indian visually impaired (VI) population.

Methods.: Data from 120 VI adults administered both instruments concurrently, were fitted to the Rasch measurement model to demonstrate that each instrument satisfies the assumptions of the model (including unidimensionality by principal components analysis); and both instruments can be cocalibrated onto a single underlying continuum of VF.

Results.: Both instruments required category reorganization for optimal rating scale functioning and possessed similar measurement precision (person separation = 2.76). Separate analysis of each instrument (eigenvalues, 2.3 and 1.9 for LFVFS39 and VFS of IND-VFQ, respectively) and the pooled 28-item analyses (eigenvalue, 2.8) satisfied the assumptions of the Rasch model, including unidimensionality. Furthermore, all items fit in the separate and pooled analyses. Separate item and person measures for each instrument correlated strongly with estimates from the pooled data (r > 0.9 for all, P < 0.0001).

Conclusions.: Both the LFVFS39 and VFS of the IND-VFQ measure the same construct, VF, and with equal measurement precision in an Indian VI population. Both instruments can be calibrated onto a single metric, thereby, enabling a comparison of their measurement range of VF.

Introduction
A myriad of patient-reported outcomes (PROs, commonly referred to as questionnaires or instruments) have been developed in the field of ophthalmology over the past 20 to 30 years. 1 The National Eye Institute Visual Functioning Questionnaire 2 (NEI-VFQ) is among the most popular PROs. Like the majority of the PROs, the NEI-VFQ has been developed predominantly for a Western population and so the content may be perceived as irrelevant by those in developing countries, such as India. By contrast, the Indian Vision Function Questionnaire (IND-VFQ) was specifically developed for relevance to an Indian population. Using CTT, both the NEI-VFQ and the IND-VFQ have been demonstrated to possess adequate psychometric properties. 24 However, Rasch analysis conducted recently on the NEI-VFQ39 5 and IND-VFQ 6,7 (albeit in different populations) reported both instruments to be flawed. Nonetheless, re-engineering resulted in the formation of a valid unidimensional measure of visual functioning (VF) for both instruments. Consequently, the corresponding scales have been labeled as the Long form Visual Functioning scale39 (LFVFS39) and Visual Functioning Scale of the IND-VFQ (VFS of the IND-VFQ). The resemblance in the names (VFS) of the two instruments—the LFVFS39 and VFS of the IND-VFQ—encourages the assumption that there is also similarity in the content and in the underlying measurement construct (visual functioning, VF). However, this assumption hasn't been tested as yet, so the question still remains as to what extent these instruments capture the same construct (if they do). 
It is important that similar-appearing instruments (at least name wise) actually do measure the same construct because only under such conditions can instruments communicate with each other. Such a concurrence in the measurement construct will facilitate the comparison of results across studies and valid conclusions can be drawn from existing data despite use of different instruments to measure the same construct. 8 For example, studies reporting people who have taken the LFVFS39 can be compared with those that have taken the VFS of the IND-VFQ. 
In addition to providing detailed insight into performance of instruments, Rasch analysis offers the benefits of possibly cocalibrating instruments after they have been demonstrated to measure the same construct. To the best of our knowledge, except for the few studies by Massof, there is paucity of literature regarding the cocalibration of ophthalmic instruments. Massof compared four VFQs (Activities of Daily Vision Scale, Visual Function Index-14, NEI-VFQ, and Visual Activities Questionnaire) administered to a low vision sample and demonstrated that all these instruments measure a single construct; however, the instruments only differed in measurement precision and accuracy. 8,9 Massof reported that all the four VFQs can be calibrated to a common measurement scale and that the measurement scale of each of the four instruments was a linear transformation of the measurement scale estimated from the merged responses of all four instruments. 8,9  
Against the above background and encouraged by the work of Massof, we were interested in investigating if the LFVFS39 and VFS of the IND-VFQ measure the same construct, VF. Given that both instruments have been propagated to measure VF, users of these instruments would expect to obtain similar estimates of patient visual functional ability from both. A comparison of these instruments in the same patient population should help prove or disprove the validity of this expectation. An opportunity thus arose for such a comparison in the present study wherein both these instruments were administered concurrently to common persons. Therefore, our primary aim was to cocalibrate the items from both the LFVFS39 and VFS of the IND-VFQ onto a single metric to determine if they did measure the same construct in a cohort of Indian visually impaired (VI) patients. If it is valid to cocalibrate the two instruments, then item measures estimated from individual Rasch assessments of the two instruments should be a linear transformation of item measures estimated from the analysis of the pooled data. So to check the validity of pooling the data, we tested the LFVFS39 and VFS of the IND-VFQ separately against Rasch model assumptions, specifically unidimensionality, in the same cohort. Given that cocalibration enables mapping the VF from both the instruments onto a single metric, our secondary aim was to compare the measurement range of VF of LFVFS39 with that of VFS of the IND-VFQ (i.e., rates of ceiling and floor effects [percentage of cases with maximum and minimum scores]). We envisage that such comparisons would assist an investigator who is interested in evaluating the VF in VI patients, in deciding which of the two instruments could be used for his or her study. 
Methods
Instruments
The NEI-VFQ was developed using focus groups of VI patients and has been extensively validated in patients with chronic eye diseases. 2,10 Therefore, the NEI-VFQ is expected to be better suited to assess the functional difficulties in patients with a range of the visual disabilities. Like the NEI-VFQ, the IND-VFQ was also developed using focus group discussions involving the VI, so its item content is suited for this population. 3 Given that the present investigation involved the VI patients (from heterogeneous causes), we chose the NEI-VFQ and IND-VFQ (i.e., Rasch versions—the LFVFS39 and VFS of IND-VFQ) for our comparisons. 
The two instruments, LFVFS39 and VFS of the IND-VFQ, have been described in detail elsewhere and have shown acceptable psychometric qualities in methodological studies using Rasch analysis.57 We have provided supplemental tables (see Supplementary Material and Supplementary Tables S1 and S2) to indicate which original items (numbers) of both instruments (i.e., NEI-VFQ39 and IND-VFQ33) were used to create these VF scales. An overview of the key features of the LFVFS39 and VFS of the IND-VFQ is provided in Table 1. The two instruments have differences (albeit minor) with respect to their length, type of response scale, and phrasing of the items. Although both instruments conceptually purport to measure VF and thus have been labeled similarly (face validity), the actual item content differs between the two instruments. For example, the LFVFS39 asks for the level of difficulty a person has in finding something on a crowded shelf, but the VFS of the IND-VFQ asks for the level of difficulty a person has in searching for things at home. While it is “visual search” that is the function being assessed in both instruments, the item on the LFVFS39 is task-specific, but the corresponding item on the VFS of the IND-VFQ is generic in nature. 
Table 1. 
 
Key Features of the LFVFS39 and the VFS of the IND-VFQ
Table 1. 
 
Key Features of the LFVFS39 and the VFS of the IND-VFQ
Characteristic Aspect No. of Items
LFVFS39
Length 15
Type of response scale 6-category Likert scale 1
5-category Likert scale 14
Category wording Excellent (1), Good (2)
Fair (3), Poor (4), Very poor (5) 1
Completely blind (6)
No difficulty at all (1) 14
A little difficulty (2)
Moderate difficulty (3)
Extreme difficulty (4)
Stopped doing this activity because of eyesight (6)
Syntax structure of items “Would you say you're…?” 1
“How much difficulty do you have…?” 4
“Because of your eyesight how much difficulty do you have…?” 9
“Wearing glasses, how much difficulty do you have…?” 1
VFS of the IND-VFQ
Length 13
Type of response scale 5-category Likert scale All
Category wording Not at all (1) All
A little (2)
Quite a bit (3)
A lot (4)
Cannot do this because of my sight (5)
Syntax structure of items “Because of your vision how much problem do you have in…?” All
We adopted rigorous procedures as recommended by the World Health Organization for translation and adaptation of instruments. 11 In brief, we translated the LFVFS39 from the source language (English) to the target language (Telugu) and then back-translated to verify accuracy. This was followed by cognitive debriefing in a representative sample to check the appropriateness of the item wording and content. Given that we used the preexisting local language versions of the IND-VFQ proposed by the developers themselves, no further cultural or linguistic adaptation of the instrument was required. Higher scores on both the LFVFS39 and VFS of the IND-VFQ indicated worse VF. 
Study Population
Both the instruments were administered concurrently by research assistants in a face-to-face interview to 120 VI patients referred to the Vision Rehabilitation Centres of L V Prasad Eye Institute (LVPEI). The two instruments were part of the packet that also included an assessment of sociodemographic and ocular condition–related data. Although unintentional, the arrangement of the instruments in the packet led to the LFVFS39 always being administered prior to the VFS of the IND-VFQ. The study was approved by the ethics committee of the LVPEI and all patients who agreed to participate signed a consent form. The research was conducted in accordance with the tenets of the Declaration of Helsinki. 
Participants ranged in age from 18 to 88 years (mean ± SD = 45 ± 17.9 years) and 79% were males. Median duration of vision loss was 6 years (range, 0.5–50.5 years). Habitual visual acuity in the better eye ranged from 20/20 to light perception (mean = 1.02 logMAR; SD = 0.54 logMAR). Retinal disorders (heredomacular degeneration, retinitis pigmentosa, diabetic retinopathy, and age-related macular degeneration) were the common diagnoses (60%) in our participants. We included participants at both extremes of the visual acuity distribution (20/20 to light perception) such that a sufficiently wide range of visual disability was represented in order to effectively evaluate the range of the instruments in our comparisons. We included participants with visual acuity of 20/20 (n = 2) because they had accompanying visual field loss in the better eye (less than 20 degrees). 
Statistical Analyses
Comparing the LFVS39 and VFS of the IND-VFQ.
We analyzed the data in two steps to answer the aims of the study. In step 1, we performed separate Rasch analysis of LFVFS39 and VFS of the IND-VFQ to determine if each satisfied the assumptions of the Rasch model (specifically unidimensionality—measuring a single construct—VF). Following confirmation of unidimensionality for each instrument in the separate analyses, in the second step we cocalibrated the items from both instruments (as if both the instruments were merged into one large instrument of 28 items) onto a single metric to determine if the pooled item set was also a unidimensional measure of VF. Cocalibration enables people's measurements on different instruments to be compared on an identical “ruler” of the construct (for example, VF). 22 In the present case, cocalibration permitted placing all the items on the same underlying continuum of VF and comparison of the relative location of the items and persons between the two instruments. Subsequently, we used the person-item map to examine the extent to which the items in the LFVFS39 covered the same or wider range of difficulty than the items in the VFS of the IND-VFQ. Also, we compared the ceiling and floor effects, targeting, and measurement precision of the pooled data set against the separate analyses of each instrument. As such, floor effect is defined as the least possible score, indicating that participants have maximum visual disability (or least VF), and ceiling effect occurs when there is maximum possible score, implying that participants have “no visual disability” (or maximum VF). 
Rasch Analysis.
Rasch analysis was performed using the Andrich rating scale model 12 with analytical software (WINSTEPS version 3.68.0; WINSTEPS, Chicago, IL), 13 to estimate interval measurement from ordinal data and the unit is logits (log-odd units). There are a series of components to Rasch analysis that have been detailed in the same journal elsewhere. 14 In brief, we assessed the following: (1) behavior of response categories; (2) measurement precision (using person separation reliability, PSR); (3) dimensionality (using fit statistics–infit MnSq criterion 0.5-1.5 1517 and principal components analysis [PCA] of residuals with criterion that the secondary dimension should have the strength of at least three items as measured by an eigenvalue >3.0 to be considered a second dimension that is greater than the magnitude seen with random data 18,19 ); (4) targeting (using a person-item map); and (5) differential item functioning (DIF; notable DIF >1.0 logit 20,21 ). 
Descriptive statistics was used to summarize demographic and clinical characteristics. P < 0.05 was chosen as the criterion for statistical significance. Analyses were performed using statistical software (SPSS software version 16.0 Windows; SPSS Inc., Chicago, IL). 
Results
Rasch Analysis of the LFVFS39 and VFS of the IND-VFQ
The LFVFS39 displayed disordered thresholds in both its response category formats. Disordering for format I is shown by the lack of monotonic increase of the thresholds (from category 3 to 4) in Table 2 (left half). A couple of rescoring options (combining category 3 and 2 vs. 3 and 4) were tested and the combination that provided the maximum measurement precision (i.e., PSR) was retained. Combining category 3 with 2 was found to be the most appropriate solution (PSR = 0.87) and there was a monotonic increase of the thresholds (Table 2, right half). Similarly, disordering of thresholds for format II is shown in Table 3 (left half). It can be seen that there is a lack of monotonic increase of the thresholds from category 2 to 3, and from 3 to 4. We initially combined categories 2 and 3 (PSR = 0.88), and the thresholds showed a monotonic increase (Table 3, right half). No further action was required. 
Table 2. 
 
Category Statistics of the Response Category Format I of the 15-item LFVFS39 before (left, 6 categories) and after (right, 5 categories) the Collapsing Procedure (n = 120)
Table 2. 
 
Category Statistics of the Response Category Format I of the 15-item LFVFS39 before (left, 6 categories) and after (right, 5 categories) the Collapsing Procedure (n = 120)
LFVFS39 Response Category Format I*
6 Categories 5 Categories†
Response Category‡ Count % Infit MnSq Threshold Category Measure Response Category Count % Infit MnSq Threshold Category Measure
1 2 2 1.90 None −4.54 1 2 2 1.49 None (−5.07)
2 8 7 1.06 −3.35 −2.67 2 17 15 0.90 −3.90 −2.82
3 9 8 1.08 −1.15 −1.64
4 49 41 1.16 −2.16 −0.60 3 49 41 1.17 −1.63 −0.71
5 49 41 1.00 0.17 2.10 4 49 41 1.04 0.14 2.09
6 3 3 1.07 4.00 5.11 5 3 3 1.08 4.00 (5.11)
Table 3. 
 
Category Statistics of the Response Category Format II of the 15-item LFVFS39 before (left, 5 categories) and after (right, 4 categories) the Collapsing Procedure (n = 120)
Table 3. 
 
Category Statistics of the Response Category Format II of the 15-item LFVFS39 before (left, 5 categories) and after (right, 4 categories) the Collapsing Procedure (n = 120)
LFVFS39 Response Category Format II*
5 Categories 4 Categories†
Response Category‡ Count % Infit MnSq Threshold Category Measure Response Category Count % Infit MnSq Threshold Category Measure
1 333 22 1.19 None −1.74 1 333 22 1.15 None (−2.24)
2 183 12 0.79 −0.11 −0.72 2 345 23 0.74 −0.88 −0.65
3 162 11 0.80 −0.11 −0.10
4 390 26 0.96 −0.62 0.61 3 390 26 0.89 −0.08 0.62
5 432 29 0.99 0.84 2.11 4 432 29 1.02 0.97 (2.28)
Like the LFVFS39, the thresholds of the categories of the VFS of the IND-VFQ also failed to show a monotonic increase, suggestive of disordering of thresholds (Table 4, left half). Here, too, a couple of rescoring options were tested (combining 2 and 3, and 3 and 4). However, combining categories 2 and 3 provided the best solution in terms of measurement precision (PSR = 0.88) and the resultant four-category format showed a monotonic increase of thresholds (Table 4, right half). 
Table 4. 
 
Category Statistics of the 13-Item VFS of the IND-VFQ before (left, 5 categories) and after (right, 4 categories) the Collapsing Procedure (n = 120)
Table 4. 
 
Category Statistics of the 13-Item VFS of the IND-VFQ before (left, 5 categories) and after (right, 4 categories) the Collapsing Procedure (n = 120)
VFS of the IND-VFQ*
5 Categories 4 Categories†
Response Category‡ Count % Infit MnSq Threshold Category Measure Response Category Count % Infit MnSq Threshold Category Measure
1 555 36 1.22 None −1.81 1 555 36 1.10 None (−2.16)
2 289 19 0.65 −0.44 −0.66 2 373 24 0.85 −0.80 −0.61
3 84 5 0.99 0.81 −0.06
4 315 20 1.14 −1.18 0.60 3 315 20 0.90 −0.01 0.60
5 315 20 1.04 0.80 2.04 4 315 20 1.03 0.82 (2.17)
Item Fit and Dimensionality
All items in the LFVFS39 as well as the VFS of the IND-VFQ displayed fit to the model. The PCA of the residuals for the LFVFS39 showed that the variance explained by the principal component was 56.4%. The unexplained variance explained by the first contrast was 2.3 eigenvalue units, providing evidence of unidimensionality in the measurement of VF by the LFVFS39. On the other hand, the PCA of the residuals for the VFS of the IND-VFQ showed that the variance explained by the principal component was 60% and the unexplained variance explained by the first contrast was 1.9 eigenvalue units. Taken together, these findings suggested that the VFS of the IND-VFQ is a unidimensional measure of VF. 
Overall Performance
The PSR was 0.88 for the LFVFS39 and VFS of the IND-VFQ, indicating that both the instruments were able to reliably discriminate among four strata of participants' VF (low, average, high, very high). Targeting was 0.33 and −0.48 logits for the LFVFS39 and VFS of the IND-VFQ, respectively, indicating the item difficulties were reasonably matched to person's VF in the sample for both the instruments (Table 5). 
Table 5. 
 
Comparison of Fit Parameters of the LFVFS39, VFS of the IND-VFQ and Pooled Data (LFVFS39 and VFS of the IND-VFQ)
Table 5. 
 
Comparison of Fit Parameters of the LFVFS39, VFS of the IND-VFQ and Pooled Data (LFVFS39 and VFS of the IND-VFQ)
Rasch Parameters LFVFS39 VFS of the IND-VFQ Pooled Data*
Number of items 15 13 28
Person separation reliability 0.88 0.88 0.94
Person separation† 2.76 2.76 3.95
Targeting (logits) 0.33 −0.48 −0.07
Differential item functioning (DIF contrast) 1 item (>1.0 logit) 0 1 item (>1 logit)
Ceiling effect (%) 5.8 13 8.3
Floor effect (%) 21.7 7.5 4.2
Differential Item Functioning (DIF)
A single item displayed notable DIF by sex and education status in the LFVFS39. The item “shaving, styling your hair, or putting on makeup” was rated was as 1.88 logits easier by females and as 1.24 logits easier by illiterates relative to other items. However, the VFS of the IND-VFQ was free from notable DIF. 
Confirmation that Both Instruments Measure the Same Construct
The pooled 28-item set fit the Rasch model well (Table 6). Single item (same as the separate analyses of the LFVFS39) showed noticeable DIF by sex and literacy status. Given that this item fit the model, it was retained. The PCA of the residuals showed that the variance explained by the principal component was 58.7%. The unexplained variance explained by the first contrast was 2.8 eigenvalue units. Taken together, these results support the unidimensionality of the pooled 28 items (like the separate assessments) from LFVFS39 and VFS of the IND-VFQ, thereby confirming that these two instruments do indeed measure the same underlying unidimensional construct of VF. 
Table 6. 
 
Item Measures for the 15-Item LFVFS39 and 13-Item VFS of the IND-VFQ Estimated from Rasch Analysis on Combined Data
Table 6. 
 
Item Measures for the 15-Item LFVFS39 and 13-Item VFS of the IND-VFQ Estimated from Rasch Analysis on Combined Data
Item No.† Item Description Item Calibration (logits) * Standard Error Infit (MNSQ)
VFS 1 Finding your way indoors 2.02 0.16 1.38
VFS 11 Going to the toilet 1.62 0.14 1.16
VFS 4 Locking or unlocking the door 1.02 0.13 1.01
LFVFS 8 Picking out and matching own clothes 0.79 0.13 1.20
VFS 3 Recognizing the face of a person standing near you 0.75 0.12 1.10
LFVFS 12 Shaving, styling your hair, or putting on makeup 0.73 0.13 1.50
VFS 5 Doing your usual work either in the house or outside 0.61 0.12 0.86
LFVFS 4 Finding something on a crowded shelf 0.40 0.12 0.59
VFS 13 Seeing the level of the container when pouring 0.35 0.12 0.87
VFS 9 Seeing differences in colors 0.21 0.12 0.89
VFS 7 Searching for things at home 0.18 0.12 1.01
VFS 8 Seeing when coming into the house after being in sunlight 0.13 0.12 0.96
LFVFS 15 Seeing and enjoying programs on TV 0.03 0.12 1.03
LFVFS 3 See well up close 0.02 0.12 1.27
VFS 10 Making out differences in coins and notes −0.04 0.12 0.89
LFVFS 7 Notice objects off to side while you are walking along −0.05 0.12 0.82
VFS 6 Doing your work up to your usual standard −0.10 0.12 1.07
VFS 12 Seeing objects that may have fallen in the food −0.28 0.12 0.97
LFVFS 13 Recognizing people you know across a room −0.35 0.12 0.90
LFVFS 6 Going down steps, stairs, or curbs in dim light or at night −0.37 0.12 0.96
LFVFS 14 Taking part in active sports or other outdoor activities that you enjoy −0.41 0.15 1.37
LFVFS 1 Eyesight using both eyes −0.50 0.15 1.19
LFVFS 9 Going out to see movies, plays, or sports events −0.58 0.16 1.08
Figures 1A and 1B illustrate the scatter plots of item measures for each instrument to estimates from the combined data. The slope was 0.99 (95% CI, 0.98 to 1.01) and the intercept was −0.34 for the LFVFS39 versus combined data (Fig. 1A). The slope was 0.94 (95% CI, 0.93 to 0.94) and the intercept was 0.39 for the VFS of the IND-VFQ versus combined data (Fig. 1B). The comparison of person measures between the two instruments demonstrated a strong linear relationship (r = 0.90, P < 0.0001). Figures 2A and 2B illustrate scatter plots of person measures estimated from the two analyses for each instrument. The slope was 1.03 (95% CI, 0.99 to 1.08) and the intercept was 0.35 for the LFVFS39 versus combined data (Fig. 2A). The slope was 1.06 (95% CI, 1.01 to 1.11) and the intercept was −0.44 for the VFS of the IND-VFQ versus combined data (Fig. 2B). Taken together, all these findings further confirm that both instruments measure the same construct. 
Figure 1. 
 
Scatter plots of item measure estimates from separate Rasch analysis. (A) 15-item LFVFS39. (B) 13-item VFS of the IND-VFQ versus measures estimated from the combined data of both the instruments. The diagonal line represents the bivariate regression line (first principal component) and the linear regression coefficients are listed in the equation on each plot. There is overlap of item measures in both the scatter plot due to which only 10 (of 15) points can be distinguished clearly for the LFVFS39 and 12 (of 13) points for the VFS of the IND-VFQ.
Figure 1. 
 
Scatter plots of item measure estimates from separate Rasch analysis. (A) 15-item LFVFS39. (B) 13-item VFS of the IND-VFQ versus measures estimated from the combined data of both the instruments. The diagonal line represents the bivariate regression line (first principal component) and the linear regression coefficients are listed in the equation on each plot. There is overlap of item measures in both the scatter plot due to which only 10 (of 15) points can be distinguished clearly for the LFVFS39 and 12 (of 13) points for the VFS of the IND-VFQ.
Figure 2. 
 
Scatter plots of person measure estimates from separate Rasch analysis. (A) 15-item LFVFS39. (B) 13-item VFS of the IND-VFQ versus measures estimated from the combined data of both the instruments. The diagonal line represents the bivariate regression line (first principal component) and the linear regression coefficients are listed in the equation on each plot.
Figure 2. 
 
Scatter plots of person measure estimates from separate Rasch analysis. (A) 15-item LFVFS39. (B) 13-item VFS of the IND-VFQ versus measures estimated from the combined data of both the instruments. The diagonal line represents the bivariate regression line (first principal component) and the linear regression coefficients are listed in the equation on each plot.
Person-item map of the cocalibrated 28 items in Figure 3 demonstrates that while the VFS of the IND-VFQ has incorporated easier items (as indicated by most items having positive logit values) that are capable of measuring lower levels of VF in patients with severe visual impairment, the LFVFS39 has relatively more difficult items. All the items of the LFVFS39 and VFS of the IND-VFQ are calibrated within 3.85 logits (range [mean ± SE] −1.83 ± 0.15 to 2.02 ± 0.16 logits). However, there are a number of items within (for example, LFVFS39 items 13 and 6) the two instruments that demonstrated similar calibrations (within 0.05 logits). The percentage of cases with ceiling and floor effects (those at the maximum and minimum scores for each instrument) are shown in Table 5
Figure 3. 
 
Person-Item map on the logit scale for the 28 combined items of the LFVFS39 and VFS of the IND-VFQ. Participants are on the left of the vertical line and those with higher visual functioning (VF) are located toward the bottom of the map. Items are located on the right of the vertical line and more difficult items are also located toward the bottom of the map. Item belonging to VFS of the IND-VFQ are shaded in grey color. Due to space constraints, item numbers of both the instruments have been provided and the correct description of items is provided in Table 6 VFS of the IND-VFQ item 1 is the least difficult item, whereas item 10 of the LFVS39 is the most difficult. Each “x” represents one participant. M, mean; S, 1SD from the mean; T, 2 SD from the mean.
Figure 3. 
 
Person-Item map on the logit scale for the 28 combined items of the LFVFS39 and VFS of the IND-VFQ. Participants are on the left of the vertical line and those with higher visual functioning (VF) are located toward the bottom of the map. Items are located on the right of the vertical line and more difficult items are also located toward the bottom of the map. Item belonging to VFS of the IND-VFQ are shaded in grey color. Due to space constraints, item numbers of both the instruments have been provided and the correct description of items is provided in Table 6 VFS of the IND-VFQ item 1 is the least difficult item, whereas item 10 of the LFVS39 is the most difficult. Each “x” represents one participant. M, mean; S, 1SD from the mean; T, 2 SD from the mean.
Table 6. 
 
Continued
Table 6. 
 
Continued
Item No. Item Description Item Calibration (logits)* Standard Error Infit (MNSQ)
LFVFS 5 Reading street signs or the names of stores −0.73 0.13 0.76
LFVFS 11 Figuring out whether bills you receive are accurate −0.91 0.13 0.83
VFS 2 Recognizing people from a distance −1.31 0.13 0.98
LFVFS 2 Reading ordinary print in newspaper −1.38 0.14 0.96
LFVFS 10 Reading small print in a telephone book, on medicine bottle, or on legal form −1.83 0.15 1.03
Discussion
We have comprehensively demonstrated via cocalibration that pooled items from the two instruments (LFVFS39 and VFS of the IND-VFQ) work well together in a consistent manner and satisfy the assumptions of the Rasch model, including unidimensionality. This provided evidence of a single underlying dominant variable and proof that they measure the same construct in the Indian VI population. 22 This finding of concordance in the measurement of the construct is significant, given the differences (albeit minor) between these instruments with regard to their item content, length, and choice of response categories. Investigators choose different instruments for a variety of reasons, including the user-friendliness of the instrument, its popularity, availability of language versions for use in a specific setting, and brevity. Consequently, this renders comparison of results of VF across studies difficult. Therefore, the demonstration of similarity of the underlying construct between instruments (two instruments in the present study) will increase our ability to compare results across studies that have used only one of these instruments. 
In an investigation of four VFQs in the low vision population, Massof reported that “if the only difference between instruments is measurement precision and accuracy, then the item and person measures estimated for each instrument from the combined data should be a linear transformation of the respective item and person measures estimated for each instrument separately.” 8 The results of the present study corroborate with those of Massof's findings regarding the measurement of a single construct by different instruments. We found a strong linear relationship between the person measure estimates of the two instruments and between the item measures of the two instruments, indicative of the fact that the person and item measures are comparable and that the same construct is being estimated by both the instruments. 
Nonetheless, both the instruments share some deficiencies: measurement redundancies (i.e., presence of majority of the items of similar difficulty, typically in the middle of the instrument) and measurement voids (i.e., a shortage of items that can differentiate among participants with different abilities, often at the difficult end of the continuum). The LFVFS39 tends to fare better as compared with the VFS of the IND-VFQ when measuring VF in those with higher levels of VF. This is perhaps because the LFVFS39 has more complex and challenging activities that involve fine resolution (e.g., reading ordinary print in newspaper, reading small print in a telephone book, and figuring out whether bills you receive are accurate) to minimize the ceiling effects. By contrast, the VFS of the IND-VFQ incorporates tasks that are too easy (yet essential) for this cohort of patients with VI for it to adequately measure the lower levels of visual disability (higher levels of VF) than the cohort possesses. Furthermore, it lacks true near-vision activities (such as those involving reading). 
The LFVFS39 is largely limited by the presence of notable DIF by sex and education level displayed by an item (shaving, styling your hair, or putting on make-up) so there could be issues about the potential bias that may be introduced under such circumstances. One of the reasons for DIF could be that this item has different activities grouped together and this has been reported as one of the causes of DIF in an instrument. 23 Participants may have found this item difficult to answer as they may have had difficulty with one task (such as putting on make-up), but not the other (such as styling hair). It is not uncommon to find items that show DIF in instruments used in health care. 23 While there are techniques to adjust for uniform DIF within the Rasch model, 24,25 it is relatively cumbersome if an instrument is to be used routinely in a clinical setting. 
Limitations of the present study are related to the relatively small sample size for a low vision rehabilitation center-based study. Given that ours was a rehabilitation center-based sample, the majority (78%) of our patients were either moderately or severely visually impaired (from nontreatable causes) and were in need of low-vision rehabilitation services. Therefore, our results may not be generalized to a moderately VI cataract sample (e.g., Mumbai). In such a population, one could expect the VFS of the IND-VFQ to demonstrate a higher ceiling effect than that displayed in the present study. Nonetheless, this needs to be investigated. It is also important to acknowledge the cross-sectional nature of this study that was not designed to assess the responsiveness or the test-retest reliability of these instruments. 
For the potential user of a VF instrument, an important question that arises is: Should the LFVFS39 or the VFS of the IND-VFQ be used for a given study? Given that both the instruments measure VF, the choice should perhaps be driven by the aim of the study. For example, if the aim is to use it as an outcome measure for low-vision interventions, then the relatively large ceiling effect of the VFS of the IND-VFQ could render it less suitable for this purpose as it will be unable to measure improvements at the lower end of the scale (i.e., in those with mild to moderate levels of visual disability). Consequently, the investigator could falsely conclude that interventions are not effective. The goal of low-vision interventions is not to improve the visual acuity or visual field per se, but to reduce the visual disability by increasing the functional reserve. 26 As the majority of low-vision devices are meant to reduce visual disability for specific distance and near tasks such as watching TV and reading, the use of the LFVFS39 may be justified given the inclusion of items on this instrument that specifically target such activities. However, if the population being investigated is a rural, illiterate, and blind population, then the VFS of the IND-VFQ could prove to be a suitable instrument—for example, to assess cataract outcomes or rehabilitation interventions such as training in mobility and activities of daily living for the incurably blind. 
Because both the LFVFS39 and VFS of the IND-VFQ seem to have their specific merits, it is very tempting to develop a new instrument embracing both the instruments. However, we believe that given the cultural and lifestyle differences across countries, the simultaneous existence of both these instruments for assessing VF in VI patients has its advantages. Future research should include constructing crosswalks between the instruments such that scores from one instrument can easily be translated to that of the other as has been done for some healthcare instruments. 2729  
In conclusion, the present study has demonstrated that both the LFVFS39 and the VFS of the IND-VFQ can be successfully cocalibrated onto the same metric, suggesting that they do indeed measure the same construct, VF, which allows for comparison of their measurement range of VF. Both the LFVFS39 and the VFS of the IND-VFQ are psychometrically robust measures of VF in Indian VI patients. 
Supplementary Materials
References
de Boer MR Moll AC de Vet HC Terwee CB Volker-Dieben HJ van Rens GH. Psychometric properties of vision-related quality of life questionnaires: a systematic review. Ophthalmic Physiol Opt . 2004;24:257–273. [CrossRef] [PubMed]
Mangione CM Lee PP Pitts J Gutierrez P Berry S Hays RD. Psychometric properties of the National Eye Institute Visual Function Questionnaire (NEI-VFQ). NEI-VFQ Field Test Investigators. Arch Ophthalmol . 1998;116:1496–1504. [CrossRef] [PubMed]
Gupta SK Viswanath K Thulasiraj RD The development of the Indian vision function questionnaire: field testing and psychometric evaluation. Br J Ophthalmol . 2005; 89:621–627. [CrossRef] [PubMed]
Murthy GV Gupta SK Thulasiraj RD Viswanath K Donoghue EM Fletcher AE. The development of the Indian vision function questionnaire: questionnaire content. Br J Ophthalmol . 2005;89:498–503. [CrossRef] [PubMed]
Pesudovs K Gothwal VK Wright T Lamoureux EL. Remediating serious flaws in the National Eye Institute Visual Function Questionnaire. J Cataract Refract Surg . 2010; 36:718–732. [CrossRef] [PubMed]
Gothwal VK Bagga DK Sumalini R. Rasch analysis of the Indian Vision Function Questionnaire. Br J Ophthalmol . 2012; 96:619–623. [CrossRef] [PubMed]
Finger RP Kupitz DG Holz FG The impact of the severity of vision loss on vision-related quality of life in India: an evaluation of the IND-VFQ-33. Invest Ophthalmol Vis Sci . 2011; 52:6081–6088. [CrossRef] [PubMed]
Massof RW. An interval-scaled scoring algorithm for visual function questionnaires. Optom Vis Sci . 2007;84:689–704. [CrossRef] [PubMed]
Massof RW. Application of stochastic measurement models to visual function rating scale questionnaires. Ophthalmic Epidemiol . 2005; 12:103–124. [CrossRef] [PubMed]
Mangione CM Berry S Spritzer K 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; 116:227–233. [PubMed]
World Health Organization. Process of translation and adaptation of instruments. In: WHO. 2007 . Available at: http://www.who.int/substance_abuse/research_tools/translation/en/. Accessed January 10, 2012.
Andrich DA. A rating scale formulation for ordered response categories. Psychometrika . 1978;43:561–573. [CrossRef]
WINSTEPS Rasch measurement [computer program]. Version 3.68.0. Chicago, IL: Winsteps; 2009.
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. [CrossRef] [PubMed]
Chien TW Lin SJ Wang WC Leung HW Lai WP Chan AL. Reliability of 95% confidence interval revealed by expected quality-of-life scores: an example of nasopharyngeal carcinoma patients after radiotherapy using EORTC QLQ-C 30. Health Qual Life Outcomes . 2010;8:68. [CrossRef] [PubMed]
Wang WC Yao G Tsai YJ Wang JD Hsieh CL. Validating, improving reliability, and estimating correlation of the four subscales in the WHOQOL-BREF using multidimensional Rasch analysis. Qual Life Res . 2006; 15:607–620. [CrossRef] [PubMed]
Pesudovs K. Item banking: a generational change in patient-reported outcome measurement. Optom Vis Sci . 2010; 87:285–293. [PubMed]
Smith EV Jr. Detecting and evaluating the impact of multidimensionality using item fit statistics and principal component analysis of residuals. J Appl Meas . 2002; 3:205–231. [PubMed]
Marella M Gothwal VK Pesudovs K Lamoureux E. Validation of the visual disability questionnaire (VDQ) in India. Optom Vis Sci . 2009; 86:E826–E835. [CrossRef] [PubMed]
Wright BD Douglas GA. Best test design and self-tailored testing. In: MESA Research Memorandum No. 19 . Chicago, IL: Statistical Laboratory, Department of Education, University of Chicago; 1975.
Wright BD Douglas GA. Rasch item analysis by hand. In: MESA Research Memorandum No. 21 . Chicago, IL: Statistical Laboratory, Department of Education, University of Chicago; 1976.
Bond TG Fox CM. Applying the Rasch Model: Fundamental Measurement in the Human Sciences . London, UK: Lawrence Erlbaum Associates; 2001.
Hambleton RK. Good practices for identifying differential item functioning. Med Care . 2006; 44:S182–S188. [CrossRef] [PubMed]
Tennant A Penta M Tesio L Assessing and adjusting for cross-cultural validity of impairment and activity limitation scales through differential item functioning within the framework of the Rasch model: the PRO-ESOR project. Med Care . 2004; 42:I37–I48. [CrossRef] [PubMed]
Langer MM Hill CD Thissen D Burwinkle TM Varni JW DeWalt DA. Item response theory detected differential item functioning between healthy and ill children in quality-of-life measures. J Clin Epidemiol . 2008; 61:268–276. [CrossRef] [PubMed]
Stelmack JA Stelmack TR Massof RW. Measuring low-vision rehabilitation outcomes with the NEI VFQ-25. Invest Ophthalmol Vis Sci . 2002; 43:2859–2868. [PubMed]
Velozo CA Kielhofner G Lai JS. The use of Rasch analysis to produce scale-free measurement of functional ability. Am J Occup Ther . 1999;53:83–90. [CrossRef] [PubMed]
Finlayson M Mallinson T Barbosa VM. Activities of daily living (ADL) and instrumental activities of daily living (IADL) items were stable over time in a longitudinal study on aging. J Clin Epidemiol . 2005; 58:338–349. [CrossRef] [PubMed]
Velozo CA Byers KL Wang YC Joseph BR. Translating measures across the continuum of care: using Rasch analysis to create a crosswalk between the Functional Independence Measure and the Minimum Data Set. J Rehabil Res Dev . 2007; 44:467–478. [CrossRef] [PubMed]
Footnotes
 Supported by the Hyderabad Eye Research Foundation, Hyderabad, India.
Footnotes
 Disclosure: V.K. Gothwal, None; S.P. Reddy, None; R. Sumalini, None; S. Bharani, None; D.K. Bagga, None
Figure 1. 
 
Scatter plots of item measure estimates from separate Rasch analysis. (A) 15-item LFVFS39. (B) 13-item VFS of the IND-VFQ versus measures estimated from the combined data of both the instruments. The diagonal line represents the bivariate regression line (first principal component) and the linear regression coefficients are listed in the equation on each plot. There is overlap of item measures in both the scatter plot due to which only 10 (of 15) points can be distinguished clearly for the LFVFS39 and 12 (of 13) points for the VFS of the IND-VFQ.
Figure 1. 
 
Scatter plots of item measure estimates from separate Rasch analysis. (A) 15-item LFVFS39. (B) 13-item VFS of the IND-VFQ versus measures estimated from the combined data of both the instruments. The diagonal line represents the bivariate regression line (first principal component) and the linear regression coefficients are listed in the equation on each plot. There is overlap of item measures in both the scatter plot due to which only 10 (of 15) points can be distinguished clearly for the LFVFS39 and 12 (of 13) points for the VFS of the IND-VFQ.
Figure 2. 
 
Scatter plots of person measure estimates from separate Rasch analysis. (A) 15-item LFVFS39. (B) 13-item VFS of the IND-VFQ versus measures estimated from the combined data of both the instruments. The diagonal line represents the bivariate regression line (first principal component) and the linear regression coefficients are listed in the equation on each plot.
Figure 2. 
 
Scatter plots of person measure estimates from separate Rasch analysis. (A) 15-item LFVFS39. (B) 13-item VFS of the IND-VFQ versus measures estimated from the combined data of both the instruments. The diagonal line represents the bivariate regression line (first principal component) and the linear regression coefficients are listed in the equation on each plot.
Figure 3. 
 
Person-Item map on the logit scale for the 28 combined items of the LFVFS39 and VFS of the IND-VFQ. Participants are on the left of the vertical line and those with higher visual functioning (VF) are located toward the bottom of the map. Items are located on the right of the vertical line and more difficult items are also located toward the bottom of the map. Item belonging to VFS of the IND-VFQ are shaded in grey color. Due to space constraints, item numbers of both the instruments have been provided and the correct description of items is provided in Table 6 VFS of the IND-VFQ item 1 is the least difficult item, whereas item 10 of the LFVS39 is the most difficult. Each “x” represents one participant. M, mean; S, 1SD from the mean; T, 2 SD from the mean.
Figure 3. 
 
Person-Item map on the logit scale for the 28 combined items of the LFVFS39 and VFS of the IND-VFQ. Participants are on the left of the vertical line and those with higher visual functioning (VF) are located toward the bottom of the map. Items are located on the right of the vertical line and more difficult items are also located toward the bottom of the map. Item belonging to VFS of the IND-VFQ are shaded in grey color. Due to space constraints, item numbers of both the instruments have been provided and the correct description of items is provided in Table 6 VFS of the IND-VFQ item 1 is the least difficult item, whereas item 10 of the LFVS39 is the most difficult. Each “x” represents one participant. M, mean; S, 1SD from the mean; T, 2 SD from the mean.
Table 1. 
 
Key Features of the LFVFS39 and the VFS of the IND-VFQ
Table 1. 
 
Key Features of the LFVFS39 and the VFS of the IND-VFQ
Characteristic Aspect No. of Items
LFVFS39
Length 15
Type of response scale 6-category Likert scale 1
5-category Likert scale 14
Category wording Excellent (1), Good (2)
Fair (3), Poor (4), Very poor (5) 1
Completely blind (6)
No difficulty at all (1) 14
A little difficulty (2)
Moderate difficulty (3)
Extreme difficulty (4)
Stopped doing this activity because of eyesight (6)
Syntax structure of items “Would you say you're…?” 1
“How much difficulty do you have…?” 4
“Because of your eyesight how much difficulty do you have…?” 9
“Wearing glasses, how much difficulty do you have…?” 1
VFS of the IND-VFQ
Length 13
Type of response scale 5-category Likert scale All
Category wording Not at all (1) All
A little (2)
Quite a bit (3)
A lot (4)
Cannot do this because of my sight (5)
Syntax structure of items “Because of your vision how much problem do you have in…?” All
Table 2. 
 
Category Statistics of the Response Category Format I of the 15-item LFVFS39 before (left, 6 categories) and after (right, 5 categories) the Collapsing Procedure (n = 120)
Table 2. 
 
Category Statistics of the Response Category Format I of the 15-item LFVFS39 before (left, 6 categories) and after (right, 5 categories) the Collapsing Procedure (n = 120)
LFVFS39 Response Category Format I*
6 Categories 5 Categories†
Response Category‡ Count % Infit MnSq Threshold Category Measure Response Category Count % Infit MnSq Threshold Category Measure
1 2 2 1.90 None −4.54 1 2 2 1.49 None (−5.07)
2 8 7 1.06 −3.35 −2.67 2 17 15 0.90 −3.90 −2.82
3 9 8 1.08 −1.15 −1.64
4 49 41 1.16 −2.16 −0.60 3 49 41 1.17 −1.63 −0.71
5 49 41 1.00 0.17 2.10 4 49 41 1.04 0.14 2.09
6 3 3 1.07 4.00 5.11 5 3 3 1.08 4.00 (5.11)
Table 3. 
 
Category Statistics of the Response Category Format II of the 15-item LFVFS39 before (left, 5 categories) and after (right, 4 categories) the Collapsing Procedure (n = 120)
Table 3. 
 
Category Statistics of the Response Category Format II of the 15-item LFVFS39 before (left, 5 categories) and after (right, 4 categories) the Collapsing Procedure (n = 120)
LFVFS39 Response Category Format II*
5 Categories 4 Categories†
Response Category‡ Count % Infit MnSq Threshold Category Measure Response Category Count % Infit MnSq Threshold Category Measure
1 333 22 1.19 None −1.74 1 333 22 1.15 None (−2.24)
2 183 12 0.79 −0.11 −0.72 2 345 23 0.74 −0.88 −0.65
3 162 11 0.80 −0.11 −0.10
4 390 26 0.96 −0.62 0.61 3 390 26 0.89 −0.08 0.62
5 432 29 0.99 0.84 2.11 4 432 29 1.02 0.97 (2.28)
Table 4. 
 
Category Statistics of the 13-Item VFS of the IND-VFQ before (left, 5 categories) and after (right, 4 categories) the Collapsing Procedure (n = 120)
Table 4. 
 
Category Statistics of the 13-Item VFS of the IND-VFQ before (left, 5 categories) and after (right, 4 categories) the Collapsing Procedure (n = 120)
VFS of the IND-VFQ*
5 Categories 4 Categories†
Response Category‡ Count % Infit MnSq Threshold Category Measure Response Category Count % Infit MnSq Threshold Category Measure
1 555 36 1.22 None −1.81 1 555 36 1.10 None (−2.16)
2 289 19 0.65 −0.44 −0.66 2 373 24 0.85 −0.80 −0.61
3 84 5 0.99 0.81 −0.06
4 315 20 1.14 −1.18 0.60 3 315 20 0.90 −0.01 0.60
5 315 20 1.04 0.80 2.04 4 315 20 1.03 0.82 (2.17)
Table 5. 
 
Comparison of Fit Parameters of the LFVFS39, VFS of the IND-VFQ and Pooled Data (LFVFS39 and VFS of the IND-VFQ)
Table 5. 
 
Comparison of Fit Parameters of the LFVFS39, VFS of the IND-VFQ and Pooled Data (LFVFS39 and VFS of the IND-VFQ)
Rasch Parameters LFVFS39 VFS of the IND-VFQ Pooled Data*
Number of items 15 13 28
Person separation reliability 0.88 0.88 0.94
Person separation† 2.76 2.76 3.95
Targeting (logits) 0.33 −0.48 −0.07
Differential item functioning (DIF contrast) 1 item (>1.0 logit) 0 1 item (>1 logit)
Ceiling effect (%) 5.8 13 8.3
Floor effect (%) 21.7 7.5 4.2
Table 6. 
 
Item Measures for the 15-Item LFVFS39 and 13-Item VFS of the IND-VFQ Estimated from Rasch Analysis on Combined Data
Table 6. 
 
Item Measures for the 15-Item LFVFS39 and 13-Item VFS of the IND-VFQ Estimated from Rasch Analysis on Combined Data
Item No.† Item Description Item Calibration (logits) * Standard Error Infit (MNSQ)
VFS 1 Finding your way indoors 2.02 0.16 1.38
VFS 11 Going to the toilet 1.62 0.14 1.16
VFS 4 Locking or unlocking the door 1.02 0.13 1.01
LFVFS 8 Picking out and matching own clothes 0.79 0.13 1.20
VFS 3 Recognizing the face of a person standing near you 0.75 0.12 1.10
LFVFS 12 Shaving, styling your hair, or putting on makeup 0.73 0.13 1.50
VFS 5 Doing your usual work either in the house or outside 0.61 0.12 0.86
LFVFS 4 Finding something on a crowded shelf 0.40 0.12 0.59
VFS 13 Seeing the level of the container when pouring 0.35 0.12 0.87
VFS 9 Seeing differences in colors 0.21 0.12 0.89
VFS 7 Searching for things at home 0.18 0.12 1.01
VFS 8 Seeing when coming into the house after being in sunlight 0.13 0.12 0.96
LFVFS 15 Seeing and enjoying programs on TV 0.03 0.12 1.03
LFVFS 3 See well up close 0.02 0.12 1.27
VFS 10 Making out differences in coins and notes −0.04 0.12 0.89
LFVFS 7 Notice objects off to side while you are walking along −0.05 0.12 0.82
VFS 6 Doing your work up to your usual standard −0.10 0.12 1.07
VFS 12 Seeing objects that may have fallen in the food −0.28 0.12 0.97
LFVFS 13 Recognizing people you know across a room −0.35 0.12 0.90
LFVFS 6 Going down steps, stairs, or curbs in dim light or at night −0.37 0.12 0.96
LFVFS 14 Taking part in active sports or other outdoor activities that you enjoy −0.41 0.15 1.37
LFVFS 1 Eyesight using both eyes −0.50 0.15 1.19
LFVFS 9 Going out to see movies, plays, or sports events −0.58 0.16 1.08
Table 6. 
 
Continued
Table 6. 
 
Continued
Item No. Item Description Item Calibration (logits)* Standard Error Infit (MNSQ)
LFVFS 5 Reading street signs or the names of stores −0.73 0.13 0.76
LFVFS 11 Figuring out whether bills you receive are accurate −0.91 0.13 0.83
VFS 2 Recognizing people from a distance −1.31 0.13 0.98
LFVFS 2 Reading ordinary print in newspaper −1.38 0.14 0.96
LFVFS 10 Reading small print in a telephone book, on medicine bottle, or on legal form −1.83 0.15 1.03
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×