Window　opening　behavior　is　a　common　and　effective　way　for　occupants　to　improve　indoor　thermal　comfort　and　air　quality.　There　are　many　studies　on　the　window　opening　behavior　in　residential　buildings,　but　research　on　infant　families　is　still　lacking.　This　paper　examines　the　window　opening　behavior　of　four　infant　families　in　Beijing　and　Tianjin　that　are　located　in　a　cold　region　of　China　during　the　transition　season.　The　Kendall　Tau　coefficient　is　used　to　measure　the　correlation　between　behavior　and　environmental　factors　for　the　first　time.　Two　types　of　windows　opening　behavior　were　found:　i)　related　to　environmental　factors　and　ii)　independent　of　environmental　factors.　The　study　reveals　that　the　window　opening　probability　in　the　living　room　of　infant　families　is　70.6%,　and　in　the　bedroom　is　57.2%,　which　is　obviously　higher　than　the　existing　research　on　ordinary　residents　(24.3　similar　to　43.2%).　The　window　opening　probability　of　infant　families　shows　a　more　obvious　downward　trend　with　increasing　outdoor　PM2.5　concentration　compared　to　ordinary　families.　Non-environmental　factors,　such　as　time　of　day　and　age　of　infant,　also　affect　behavior.　Based　on　the　proposed　concept　of　indoor　CO2　driving　strength　(DSCO2_In),　it　was　concluded　that　the　variation　of　the　indoor　CO2　concentration　is　the　result　of　the　window　opening　behavior　rather　than　a　driving　factor.　The　logistic　regression　model　shows　that　outdoor　temperature　and　PM2.5　concentration　are　the　most　important　factors.　This　study　can　help　us　gain　a　thorough　understanding　of　the　window　opening　behavior　in　infant　families　and　it　can　also　provide　data　and　theoretical　support　for　of　the　simulation　of　energy　consumption　in　infant　families.　(C)　2021　Elsevier　B.V.　All　rights　reserved.