• combining a lot of studies (usually 25 or more)
  • a great place to start when digging into new topics
    • did someone already do the work? so I don’t have to
  • discovers biases &
    • e.g. heterogenuitey not a great thing to sink money into
  • different measurement problems can cancel each other out