Stupid Question 75 (Öredev conference special): Are all pairs good in RE to knowledge level in pair programming?

[To celebrate my first year of programming I will ask a ‘stupid’ questions daily on my blog for a year, to make sure I learn at least 365 new things during my second year as a developer]

These questions are from my session at Öredev 2012: Stupid questions and generation n00b: top ten intriguing things you should do

All the posts related to my session can be found here: Öredev related

pair programming

I’m always interested in how I can improve my learning and my skills as a developer, so when I was first introduced to the concept of pair programming I was very curious. Except from wondering if this actually works, as I have had both bad and good experiences and I haven’t evaluated the effectiveness or the quality, I was also wondering if all combinations are good when it comes to knowledge.

Finding research on this was hard, but I managed to find some. And what is very interesting is that the conclusions of the study explains a lot what I have seen in school, and at work. Basically a pairs tend to have a better outcome more when it’s novice-novice (junior-junior), or/and if the one or both of the programmers are presented with something they don’t know, such as a new language, framework or problem (or one of them is a novice). But, the effectiveness findings were inconsistent as is normal with studies of pair programming, suggesting that pair programming isn’t guaranteed to be successful even if the combination has a novice - but the likelihood increases. Obviously we need more studies, but this is an interesting start!

You can find one interesting study here: Pair programming productivity: novice-novice vs expert-expert
And here : The effectiveness of pair programming: A meta-analysis
As for what makes novice-novice pairs work, I will get to that in the next question: Stupid Question 76: Are there any guidelines for junior developer pair-programming?

Comments

Last modified on 2012-11-05