So let me see if I've got this right: Alma Adams was pegged as Mel Watt's replacement when his term was cut short by his appointment to Director of FHFA, but instead of having the election for his replacement at that time there was a retroactive vote for what Adams had already served as well as another vote for if she would continue. Is that right?
On a side not, District 9 did not have a Democratic challenger that year. How would you suggest addressing this? Leaving the district in just makes it seems like it was skewed in favor of the Democrats since they wasted 0 votes and republicans wasted 50%, but in reality it likely would have shown another Republican favoring gerrymander had any Democrat decided to run against the Republican there.
To the first question: I would not say Alma Adams was literally selected as the replacement as soon as Watt resigned; House members cannot be appointed, and our favorite LGBT advocate Pat McCory would certainly not have chosen a Democrat to replace Watt anyway
. With the district being so heavily Democratic, though, she was sure to win the special and general elections in November as long as she won the Democratic primary earlier in the year. Also, House terms last for two years. After the special election on 11/4/14, Adams was immediately sworn into Watt's former seat for the remaining two months of the unexpired term; there was simply a vacancy from the period of Watt's resignation to Adams' swearing-in. No "retroactive" election was held for a representative during this time period, as you cannot be considered a member of Congress before you have been elected.
To the second question: I would probably take in that data as it is, and leave a note that there was no Democratic challenger. Knowing North Carolina, it is quite likely that there will be some startling results when the efficiency gap is applied on its elections.