Question

我在理解这个实验是交叉的还是嵌套的分组因子时遇到了问题，我想与LMM中的随机效应相关联。我也想了解模型的语法是否正确。

您将如何处理：该实验设置了三个block复制。其中有三个treatment地块（这些是土壤地块，总共9个）。在每个图中，我有六个点（observations-这代表我的土壤剖面），我在其中收集土壤样本。因此，在observations中，我有三个土壤horizons（这三层土壤来自同一点-observation，但深度不同）。

我主要想测量同一土壤treatments中horizon之间的差异。

one_hor是一个缩小的数据帧，其中仅包含一个地平线的mydependvariable：这是数据帧：

dput(one_hor)
structure(list(myanagem = structure(c(1L, 2L, 3L, 1L, 2L, 3L, 
1L, 2L, 3L), .Label = c("CH", "CHF", "WTH"), class = "factor"), 
    block = structure(c(1L, 1L, 1L, 2L, 2L, 2L, 3L, 3L, 3L), .Label = c("1", 
    "2", "3"), class = "factor"), hor = structure(c(3L, 3L, 3L, 
    3L, 3L, 3L, 3L, 3L, 3L), .Label = c("F", "H", "L"), class = "factor"), 
    mydependvariable = c(47.8441491508198, 42.9776016786983, 
    44.2640639539627, 41.5914174334551, 51.6427380254114, 37.5819989709135, 
    49.8987347045142, 41.3728234591754, 52.9627860231326)), row.names = c(NA, 
-9L), class = "data.frame")

为此，我想到了以下模型：

model_lmer_C_L <- lmer(log(mydependvariable) ~  managem + (1|block),
       data= mix_L, REML = TRUE)

我还想测量同一horizons中的土壤treatment之间的差异（我不知道这是否满足独立性的假设）。为此，我使用了数据框all_hors，该数据框具有各个角度的值：

dput (all_hors)
structure(list(myanagem = structure(c(1L, 2L, 3L, 1L, 2L, 3L, 
1L, 2L, 3L, 1L, 2L, 3L, 1L, 2L, 3L, 1L, 2L, 3L, 1L, 2L, 3L, 1L, 
2L, 3L, 1L, 2L, 3L), .Label = c("CH", "CHF", "WTH"), class = "factor"), 
    block = structure(c(1L, 1L, 1L, 2L, 2L, 2L, 3L, 3L, 3L, 1L, 
    1L, 1L, 2L, 2L, 2L, 3L, 3L, 3L, 1L, 1L, 1L, 2L, 2L, 2L, 3L, 
    3L, 3L), .Label = c("1", "2", "3"), class = "factor"), hor = structure(c(3L, 
    3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 
    3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L), class = "factor", .Label = c("F", 
    "H", "L")), mydependvariable = c(44.6968250852355, 53.1828782803165, 
    47.4575645657112, 51.5093038256725, 48.1827652837951, 39.9179043777532, 
    50.0989615281406, 41.7977368903542, 43.0292001919997, 50.4325507695009, 
    49.7646553556968, 53.9195107181261, 45.2323395692346, 52.0066022471402, 
    54.3295196086515, 41.7231684901803, 51.211617259873, 48.1669047180151, 
    56.9066150534588, 42.1880849195177, 67.6490806984713, 62.2504530592383, 
    43.5937599989346, 39.4377478805407, 69.1207400428221, 36.5530072488604, 
    42.3732207709937)), row.names = c(NA, -27L), class = "data.frame")

为此，我想到了以下模型：

model_lmer_C <- lmer(log(mydependvariable) ~  treatment * horizon + (1|block/treatment), data = all_hors)

您认为模型的语法正确吗？

我的理解是“治疗”和“障碍”是交叉的，对吗？

我不确定土壤“水平”是否嵌套或交叉在“观测”中（也许只有在我想测量同一“处理”中土壤“水平”之间的差异时才如此）？ / p>

具有随机效应的LMM与交叉/嵌套分组因子的水平有关

0 个答案: