I suggest that you make it in spss have the option to reverse the items and save your data and pass it to the PLS.

Hello Asif,

I vote also for recoding the scores on these items, and interpreting the estimated relationships in light of the reversed polarity of the wording.

Here's a prior RGate thread on the same question: https://www.researchgate.net/post/How_to_analyse_questionnaire_with_Negative_item_reverse_item_and_positive_item

You may stop here, or travel a bit further along this explanation...

There is abundant evidence that, in survey research, "opposite" or "reverse" worded stimuli do not yield arithmetic complement scores, so it makes more sense to honor the wording of the item/question/stimulus as it was.

Example: "I love the color RED" and "I hate the color RED" may appear to be logical opposites, but will almost never yield complementary scores (e.g., using a Likert-type response scale of 1 = strongly disagree, 2 = disagree, 3 = neither agree nor disagree, 4 = agree, 5 = strongly agree) such that, if the first stimulus (love RED) yielded a median score of 3.5, the second stimulus will likely have a median different from 2.5.

One of the usual reasons for reversing wording is to avoid response bias, but here's an interesting study that suggests it may be fruitless: Article Ineffectiveness of Reverse Wording of Questionnaire Items: L...

This link indicates that estimated factor structures can be affected by reverse worded stimuli: Article Examining the Effect of Reverse Worded Items on the Factor S...

As well: http://www.psicothema.com/pdf/4463.pdf

And, finally: https://pdfs.semanticscholar.org/f514/437ec8f08aa42a83e2025e5efbe000039b37.pdf

Good luck with your work.

Thank you David Morse Cristian Ramos-Vera