I'm a lurker who is very much indebted to the intelligent, articulate and tenacious GC posters on FWR for opening my eyes to the whole TRA can of worms.
I've also conducted and peer-reviews behavioural research in human subjects for almost two decades as part of my job. Even leaving aside its sociocultural context and small sample size, that Iranian study has significant methodological flaws that undermine the conclusions that can be drawn from the study.
For starters, it's a cross-sectional study rather than a longitudinal one. The participants in the pre-op and post-op study groups aren't the same people followed over time as they undergo surgery, they're different groups of people. Therefore, strictly speaking, you can't conclude from these data that trans individuals become happier as a result of surgical intervention; just that, on average, the post-op trans people in this study were happier than the pre-op trans people in this study.
Whether this is generalisable to the Iranian trans population as a whole (let alone the rest of the world) depends on how representative the study population was of the Iranian trans community. The recruitment process used for this study was open to sampling bias. Participants for this study were so difficult to recruit that they used a "snowball" sampling method; "one individual was selected by the researcher and s/he was asked for his/her friends who were in the same situation". It isn't clear whether participants may have interpreted "in the same situation" to mean "another post-op trans person" or "another post-op trans person who is as happy as me after surgery". Whether individuals with a negative outcome from surgery are likely to be in social contact with those whose outcome was more positive is open to speculation.
The results of a statistical analysis on the "happiness" data are reported, but not the actual data - even though the authors do report the data for the mental health scales they used for the second part of the analysis. The most common reason for this sort of scientific "fudge" is that the raw data show something the authors don't want to report - in this case, perhaps that, despite the significant difference between the happiness ratings of the pre-op and post-op groups, the ratings weren't particularly high in either group...?
Not forgetting that most of the Introduction is a spectacular conflation of sex and gender, (even allowing for ESL) resulting in such statements as:
The majority of people have either the characteristics of masculinity or femininity, but some have the characteristics of both men and women (3) called transsexual people.
...gender is determined by physiological, biological and anatomical differences genetically determined at birth.