HTML-encoded characters in news article RSS feeds can easily be removed by re-editing the news article in "html editing mode".
When a news module is exposed via RSS, the summary field of each news article appears as the description field of each RSS item. Sometimes this description field contains html-encoded characters e.g. &lsquot; (“).
The html-encoded characters have been inserted by the rich text editor that was used to create the news articles in the News Module admin. Some symbols are always saved as html-encoded characters, whether they are copied in from another application (e.g. MS Word) or whether entered directly into the editor from the keyboard.
This doesn't cause any problems for the most part because the news articles are generally displayed in browsers which automatically decode them.
Unfortunately, other applications may consume & display the RSS feed (e.g. a flash movie), which may not automatically decode them.
The best way to work around this issue is to re-edit each problematic news article using "html editing mode".
There is no need to actually edit any fields - purely by saving the summary field from a text area (as opposed to the rich text editor) will remove any html-encodings and replace them with the relevant symbol character itself.