Premium
Validation of two brief questionnaires designed to assess fruit and vegetable intake among 3 rd –5 th grade SNAP‐Ed participants
Author(s) -
Zhang Youjie,
Hurtado Ghaffar,
Reicks Marla
Publication year - 2016
Publication title -
the faseb journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.709
H-Index - 277
eISSN - 1530-6860
pISSN - 0892-6638
DOI - 10.1096/fasebj.30.1_supplement.896.9
Subject(s) - wilcoxon signed rank test , test (biology) , medicine , mathematics , demography , statistics , mann–whitney u test , biology , paleontology , sociology
Reliable, valid and easy‐to‐implement tools are needed to monitor behavior change among youth SNAP‐Ed participants. The Food Web questionnaire (FWQ) and the Plate Activity questionnaire (PAQ) are two brief instruments designed to assess fruit and vegetable intake among 3 rd –5 th grade children. Children were asked to write eight foods that they ate yesterday in the FWQ and to indicate the proportion of their dinner plate covered with fruits and vegetables the previous night in the PAQ. To examine test‐retest reliability, children (n=27) completed the FWQ and PAQ on two occasions, one week apart. Children also took photos of their dinner plate on both occasions. Wilcoxon signed‐rank test showed that the numbers of total foods, fruits and vegetables and the PAQ scores remained stable between test‐1 and test‐2, respectively. To determine convergent validity, children (n=65) completed the FWQ and PAQ followed by an interviewer‐administered 24 hour dietary recall (24HR). Although Wilcoxon signed‐rank tests didn't show significant differences in terms of counts of fruits and vegetables between the FWQ and 24HR, the match rates of fruit and vegetable items were relatively low (38.3% and 41.3%, respectively) and the omission and intrusion rates were relatively high (38.3% and 23.4%, and 38.7% and 20.0%, respectively). Pearson correlation analysis showed a positive relationship between the PAQ scores and 24HR dinner fruit and vegetable servings (r=0.31, p<0.001). However, fruit and vegetable intake at dinner was not significantly different by PAQ scores (F (4, 60) =2.25, p=0.074). Of the 36 cases of researcher‐assigned PAQ scores based on dinner plate photos taken by the participants, only 1/3 of participants’ scores matched researcher‐assigned scores, with more overestimations than underestimations (42% vs 25%). In conclusion, test‐retest reliability was acceptable for the FWQ and the PAQ, but convergent validity was questionable within the current sample. Revisions of the questionnaires may be necessary to improve convergent validity. Support or Funding Information University of Minnesota Extension SNAP‐Ed