Appropriate item aggregation for one construct

Hi all,

I have just finished collecting my data and I have a question concerning how to deal with a construct that I measured. In the questionnaire participants were asked to name at least one and a maximum of three persons which they follow on social media. After that, they had to answer questions concerning the likeability of the person(s) which they named. The construct was then measured with the person that scored the highest on likeability if more than one persons were named. Now I have the construct measured three times in my data. About 200 times for the first named person, around 70 for the second named and 40 for the third one. The construct consists of 16 items but now I have 48 items. Is it okay to add up the single items of the three choices and after that calculate the mean to get one construct with which I can proceed the analysis (factor analysis, ANOVA, etc.)? Theoretically they all have in common that they are the most liked person of the participant. It’s not about a repeated measure nor time series data and I don’t want to compare the three groups, i’d rather have one construct but I can’t find an answer for my problem.

Thanks in advance for your support!