You may try children pictures about text ( see Eisner,1972).

See also Lowenfeld & Btittain, 1975.

When I worked at a literacy clinic we used Woods & Moe ARI. It's better one-on-one though. For groups, I would use the Burke reading inventory or a cloze procedure.

Hope this helps.

Woods, M. L., & Moe, A. J. (2007). Analytical reading inventory : comprehensive standards-based assessment for all students including gifted and talented (8th ed.). Upper Saddle River, N.J.: Pearson/Merrill Prentice-Hall.

Dear Amy,

Previous answers provide you with some ways of measure text comprehension in children.

If I were you I would appeal to the following procedure:

1. To have some vignettes with sentences that contain key ideas of the text at hand. May be three vignettes.

2. To have other vignettes that contain ideas of the text, but that they are not key or central ideas of the test. May be three vignettes. 

3. To have other vignettes that have nothing to do with the text at hand. May be three vignettes.

After this, I would first put all these vignettes on a table at the reach of the child and then I would ask the child to give me (a) the vignettes s/he thinks best describe what is more important in the text and why; (b) those vignettes that refer to the text but that do not describe what is more important in the text and why; and (c) those vignettes that having nothing to do with the text at hand and why? In order to avoid an effect order, I would counterbalance (a), (b), and (c) across children. Finally, I would see how many children chose or did not choose the appropriate vignettes and if their justifications for their choices were consistent or inconsistent with the mentioned choices or answers.  

Of course the number of vignettes can vary according to the length and degree of difficulty of the focal text.

I hope that this helps you.

Best regards