Printed questionnaire are more difficult to be collected. In fact, it was almost impossible to collect them in a recent survey. Use of tablets (or similar devices) and more appealling formats (i.e. colours) might work as they limit completion time and are more enjoyable. We found a shop's' cafeteria as the best place to collect data. Type of shop and culture also influence data collection.
Take a look at the Nordstrom Innovation Lab (https://datafloq.com/read/how-fashion-retailer-nordstrom-drives-with-innovat/398). Keep in mind the lab has since been a victim of corporate restructuring (http://www.geekwire.com/2015/nordstrom-shrinks-innovation-lab-reassigns-employees-shakeup-tech-intiatives/) - but it's an interesting experiment about getting results from customers and learning from co-creating with them. There's a role for traditional market research and UX tools (especially participatory design) but maybe even those are evolving, and more real time efforts like Nordstrom's point to an evolution in research technique's. Agree with Nicholas' points - also would well-placed kiosks with incentives to participate ('fill out the survey, take 20% off your bill if you're still shopping') work as opposed to the use of tablets? Something to think about.