Google N-gram Corpus Counts
| 子供じみた　 | 35,666 | 80.8% |
| 子どもじみた |  5,865 | 13.3% | <- adding
| 子供染みた　 |  2,063 |  4.7% |
| 子ども染みた |    235 |  0.5% |
| こどもじみた |    326 |  0.7% |

@@ -5,0 +6,3 @@
+</k_ele>
+<k_ele>
+<keb>子どもじみた</keb>

I don't think the x-ref is needed.

@@ -15 +14,0 @@
-<xref type="see" seq="2103380">子どもっぽい・こどもっぽい</xref>
@@ -17 +16 @@
-<gloss>child-like</gloss>
+<gloss>childlike</gloss>

Looking at those numbers I'd leave it alone.

@@ -14,0 +15 @@
+<xref type="see" seq="2103380">子どもっぽい・こどもっぽい</xref>
@@ -15,0 +17,3 @@
+<gloss>child-like</gloss>
+<gloss>immature</gloss>
+<gloss>infantile</gloss>

daij both only have 子供染みる (and I just added an entry 
for おとなじみる), but usage is mostly as an adj, in the 
past sense.
子供じみた	1915
子供染みた	72
子供じみる	3
子供染みる	No matches
There's also a sound file associated with this entry. 
Should we change it to ～じみる, add a new entry for ～じ
みる or change 大人じみる etc. to ～じみた？