Hello fellow academics.

So i have been struggling with a very low cronbach's alpha on a sub scale i.e alpha of .383 to be precise. After following the advice of some academics on research gate i analysed the inter item correlations and looked at the "alpha if item removed" column. It appeared that there is a specific item that was causing the alpha to be so low and after removing the item, the cronbach's alpha went up to .618.

My question is... because i have removed the item when calculating the reliability does this mean i should also remove the item from the composite scale when running other tests such as descriptive, correlations etc or is it OK to include the item for other analyses and remove it only when calculating the alpha?

Thanks is advance!