Dataframe format

Packages: dplyr, tidyr


To explore or analyze the data it is sometimes convenient to change its format, meaning how the information is structured in the dataframe. The usual formats are “long” and “wide” formats. But other formats can be useful. Sometimes, you’ll need to aggregate the data such at the day level or the participant-level. In the context of dyads, there are also specific formats that can be useful, in particular one in which partners’ variables are displayed side by side.

Wide format

In a wide format, each participant has a unique row of values. Each time point and variable combination gives a column (e.g., PA1_time1, PA1_time2, PA2_time1, PA2_time2).

The original dataset has a long format, as follows:

  dyad role obsno id age cond_dyad           scheduled
1    1    1     1  1  40     condB 2018-10-17 08:00:08
2    1    1     2  1  40     condB 2018-10-17 09:00:01
3    1    1     3  1  40     condB 2018-10-17 09:59:56
4    1    1     4  1  40     condB 2018-10-17 10:59:48
5    1    1     5  1  40     condB 2018-10-17 12:00:12
6    1    1     6  1  40     condB 2018-10-18 07:59:47
                 sent               start
1 2018-10-17 08:00:11                <NA>
2 2018-10-17 09:00:22                <NA>
3 2018-10-17 10:00:08                <NA>
4 2018-10-17 10:59:52 2018-10-17 11:00:12
5 2018-10-17 12:00:15                <NA>
6 2018-10-18 08:00:08                <NA>
                  end contact PA1 PA2 PA3 NA1 NA2 NA3
1                <NA>      NA  NA  NA  NA  NA  NA  NA
2                <NA>      NA  NA  NA  NA  NA  NA  NA
3                <NA>      NA  NA  NA  NA  NA  NA  NA
4 2018-10-17 11:03:01       0   1  11  25  10  16  28
5                <NA>      NA  NA  NA  NA  NA  NA  NA
6                <NA>      NA  NA  NA  NA  NA  NA  NA
  location daycum valid_var valid daycum_dyad beepno
1     <NA> 1 days         1     0      1 days      1
2     <NA> 1 days         1     0      1 days      2
3     <NA> 1 days         1     0      1 days      3
4        A 1 days         7     1      1 days      4
5     <NA> 1 days         1     0      1 days      5
6     <NA> 2 days         1     0      2 days      1

To reformat you data from a long format, you first have to select a subset including the nesting variable (e.g., ‘id’), the ‘obsno’ variable (indicate the order of the beeps), and the variables of interest (e.g., ‘PA1’, ‘PA2’). Then, use the built-in function ‘reshape()’ as follows:

# Select
df = data[,c("id", "obsno", "PA1", "PA2")]
# Reshape
df_wide = reshape(as.data.frame(df), direction = "wide", idvar = "id", timevar = "obsno", sep="_")

The outcome dataframe looks like:

    id PA1_1 PA2_1 PA1_2 PA2_2 PA1_3 PA2_3 PA1_4 PA2_4
1    1    NA    NA    NA    NA    NA    NA     1    11
71   2    32     3    23     4    17     7    24    12
141  3     8    21    19    35    NA    NA    NA    NA
211  4    NA    NA    15    27    18    34    19    31
281  5    NA    NA    NA    NA    NA    NA    11    59
351  6    NA    NA     1     1    NA    NA     1     1
    PA1_5 PA2_5 PA1_6 PA2_6 PA1_7 PA2_7 PA1_8 PA2_8 PA1_9
1      NA    NA    NA    NA     1     1     1     1     1
71     32     6    26     1    15     1    19     1    21
141     9    13    22    25    NA    NA    24    27    12
211    18    31    NA    NA    17    29    13    20    20
281    32    56    46    68    NA    NA    36    68    NA
351    NA    NA    11     1    NA    NA    11    14     7
    PA2_9 PA1_10 PA2_10 PA1_11 PA2_11 PA1_12 PA2_12 PA1_13
1       1     NA     NA     NA     NA      1      1      1
71      2     11      2     11      1     13      1     13
141    21     NA     NA     13     30     36     57     15
211    29     20     24     NA     NA     NA     NA     33
281    NA      1     61     NA     NA     NA     NA      1
351    25     NA     NA      1     25     NA     NA      1
    PA2_13 PA1_14 PA2_14 PA1_15 PA2_15 PA1_16 PA2_16 PA1_17
1        1      1      1      1      1     NA     NA     NA
71       1      8      2     NA     NA     14      4     NA
141     30     NA     NA     NA     NA     38     48     NA
211     36     32     22     37     32     NA     NA     32
281     38     NA     NA     NA     NA     NA     NA      1
351      3      1      1      1      1     NA     NA      1
    PA2_17 PA1_18 PA2_18 PA1_19 PA2_19 PA1_20 PA2_20 PA1_21
1       NA     NA     NA     NA     NA     NA     NA     NA
71      NA      9      4     13     10     NA     NA     NA
141     NA     NA     NA     NA     NA     NA     NA     27
211     36     27     30     31     33     NA     NA     31
281     47      8     56     NA     NA     NA     NA     NA
351      1      2      1     NA     NA      9      1     11
    PA2_21 PA1_22 PA2_22 PA1_23 PA2_23 PA1_24 PA2_24 PA1_25
1       NA     NA     NA     NA     NA     NA     NA     23
71      NA     19      1     NA     NA     25     16     32
141     39     24     36     NA     NA     20     27     18
211     30     29     32     NA     NA     NA     NA     21
281     NA     NA     NA      4     64      1     70      1
351      1     NA     NA      2     20      1     22      1
    PA2_25 PA1_26 PA2_26 PA1_27 PA2_27 PA1_28 PA2_28 PA1_29
1       22     NA     NA     14      6      7      5      1
71      15     NA     NA     NA     NA     29      5     31
141     28     NA     NA     NA     NA     21     44     NA
211     30     21     31     22     37     NA     NA     27
281     50      1     57     NA     NA     NA     NA      1
351     14     NA     NA     NA     NA      1      1      1
    PA2_29 PA1_30 PA2_30 PA1_31 PA2_31 PA1_32 PA2_32 PA1_33
1        8     NA     NA      1      1      1      1      1
71       6     31      7     NA     NA     NA     NA     26
141     NA     NA     NA     NA     NA     NA     NA     14
211     33     25     26     23     33     NA     NA     25
281     41     NA     NA      1     50     NA     NA     33
351      1      1      1     NA     NA     NA     NA      7
    PA2_33 PA1_34 PA2_34 PA1_35 PA2_35 PA1_36 PA2_36 PA1_37
1        1      1      1      1      1     NA     NA      1
71       1     26      1     17      1      3      4      4
141     31     27     54     NA     NA     NA     NA     NA
211     33     NA     NA     28     37     27     30     28
281     65     NA     NA     NA     NA     NA     NA      3
351      1      9      1     10      4      8     17     NA
    PA2_37 PA1_38 PA2_38 PA1_39 PA2_39 PA1_40 PA2_40 PA1_41
1        1      1      3     NA     NA      1      1     NA
71       5     20      1     15      1      1      1     NA
141     NA     28     28     NA     NA     28     33      9
211     32     26     22     24     28     18     22     19
281     60     NA     NA     NA     NA      1     45     NA
351     NA     NA     NA     NA     NA      1      7     NA
    PA2_41 PA1_42 PA2_42 PA1_43 PA2_43 PA1_44 PA2_44 PA1_45
1       NA      1     19     NA     NA      1      1      1
71      NA     13     11     17      6      5      3      6
141     33     NA     NA     NA     NA     25     43     13
211     33     19     28     24     38     26     29     30
281     NA      1     41     NA     NA     NA     NA     NA
351     NA      1      1     NA     NA      1      1      1
    PA2_45 PA1_46 PA2_46 PA1_47 PA2_47 PA1_48 PA2_48 PA1_49
1        9      6     29     NA     NA     NA     NA     26
71       4     22      5     24      6     14      8     24
141     12     NA     NA     NA     NA     28     43     NA
211     34     NA     NA     NA     NA     21     24     26
281     NA     NA     NA     29     66     NA     NA     36
351      1     NA     NA     NA     NA     NA     NA     NA
    PA2_49 PA1_50 PA2_50 PA1_51 PA2_51 PA1_52 PA2_52 PA1_53
1       22     24     21     24     15     24      7     24
71      11     NA     NA     33      1     24      5     23
141     NA     26     52     NA     NA     27     34     NA
211     33     27     31     NA     NA     29     30     31
281     70     20     62     NA     NA      1     60      1
351     NA      6     12      1     21     NA     NA      1
    PA2_53 PA1_54 PA2_54 PA1_55 PA2_55 PA1_56 PA2_56 PA1_57
1       12     21     23     NA     NA     NA     NA      1
71      12     30     14     38     11     NA     NA     24
141     NA     NA     NA     NA     NA     23     38     14
211     29     25     26     NA     NA     NA     NA     22
281     50     NA     NA     NA     NA      1     41      1
351     15      1      5      1      1     NA     NA      1
    PA2_57 PA1_58 PA2_58 PA1_59 PA2_59 PA1_60 PA2_60 PA1_61
1        1      1      3     NA     NA      1      1      1
71       3     NA     NA     NA     NA     19      1     NA
141     40     18     37     17     32     NA     NA     NA
211     35     22     32     NA     NA     25     29     24
281     42      1     44     NA     NA     14     54     NA
351      1     NA     NA      1      1     NA     NA     NA
    PA2_61 PA1_62 PA2_62 PA1_63 PA2_63 PA1_64 PA2_64 PA1_65
1        1      1      1     NA     NA      1      1     NA
71      NA     19      4     16      1     10      1     NA
141     NA     16     14     NA     NA     13     25     26
211     26     21     29     20     22     21     31     20
281     NA     26     70     20     64     NA     NA     NA
351     NA      6      1      6      2      1     12      1
    PA2_65 PA1_66 PA2_66 PA1_67 PA2_67 PA1_68 PA2_68 PA1_69
1       NA      1      1      1      1      1      1      1
71      NA     15      1      8      1      7      3     16
141     49     14     43     NA     NA     17     35     25
211     25     NA     NA     27     30     30     35     28
281     NA     NA     NA     NA     NA     NA     NA      1
351     15     NA     NA     NA     NA     NA     NA     NA
    PA2_69 PA1_70 PA2_70
1        5      1      1
71       1     14      1
141     28     16     11
211     28     26     29
281     40      1     43
351     NA     NA     NA

Long format

Based on the previous wide format, we want to reformat this dataframe into a long format. We propose two solutions:

  • ‘reshape()’ function, a base R function
  • ‘pivot_longer()’ function from the tidyr package

# Recreate the list of time values and variable_time (names of the columns in the wide format)
time = as.character(c(1:70)) # 70 beeps per participant
list_var = lapply(c("PA1","PA2"), function(x) paste0(x,"_", time)) # PA1_1, PA1_2, ...

# Use the reshape
df_long = reshape(
  df_wide,
  idvar = "id",
  varying = list_var,
  v.names = c("PA1", "PA2"),
  timevar = "obsno",
  times = time,
  direction = "long"
)

The outcome dataframe looks like:

# A tibble: 6 × 4
     id obsno   PA1   PA2
  <dbl> <chr> <int> <int>
1     1 1        NA    NA
2     1 2        NA    NA
3     1 3        NA    NA
4     1 4         1    11
5     1 5        NA    NA
6     1 6        NA    NA

Agregate formats

For preprocessing checks or later data analysis, we may need to aggregate the data at the day level or the participant level, from an original dataset that is at the beep level. To do so, we mainly use the ‘group_by()’ and the ‘summarise()’ functions from the dplyr packages. The general idea is to:

  1. To change the dataframe level with the ‘group_by()’ function: identification and time variable can be used to handle the level of the dataframe. In addition, we can add participant-level or day-level invariant variables to keep them in the dataframe outcomes.
  2. To compute aggregate scores from the time-varying variables with the ‘summarise()’ function: mean, max, etc. of the variables of interest to aggretate them at the chosen level.

Day level

To create a dataset at the day level, we need the daycum variable. Then, ‘group_by()’ the minimal set of following variables (e.g., id and daycum). Finally, aggregate the variables of interest at the day level.

df_day = data %>%
    group_by(id, role, age, cond_dyad, daycum) %>%    # Day-level through daycum variable 
    summarise(PA1_mean = mean(PA1, na.rm=TRUE))

The outcome dataframe looks like:

# A tibble: 6 × 6
# Groups:   id, role, age, cond_dyad [1]
     id  role   age cond_dyad daycum PA1_mean
  <dbl> <int> <int> <chr>     <drtn>    <dbl>
1     1     1    40 condB     1 days     1   
2     1     1    40 condB     2 days     1   
3     1     1    40 condB     3 days     1   
4     1     1    40 condB     4 days   NaN   
5     1     1    40 condB     5 days    23   
6     1     1    40 condB     6 days     7.33

Participant level

To create a dataframe at the participant level (only one row per participant), we only need to ‘group_by()’ the id variable. Time-varying variables can be aggregated using the ‘summarise()’ function.

df_participant = data %>%
    group_by(id, role, age, cond_dyad) %>%
    summarise(PA1_mean = mean(PA1, na.rm=TRUE))

The outcome dataframe looks like:

# A tibble: 6 × 5
# Groups:   id, role, age [6]
     id  role   age cond_dyad PA1_mean
  <dbl> <int> <int> <chr>        <dbl>
1     1     1    40 condB         5.46
2     2     2    42 condB        18.5 
3     3     1    25 condB        20.3 
4     4     2    25 condB        24.4 
5     5     1    25 condB        10.8 
6     6     2    25 condB         3.39

Formats for dyadic datasets

When manipulating dyadic data, it is often necessary to reformat the dataset from an individual to a dyad format. A dyad format displays partners’ observations side by side, meaning that the first partner’s data will have its variables on the left part of the dataframe and the second partner on the right (with two options).

Individuals
A A A A A A A
A A A A A A A
A A A A A A A
A A A A A A A
B B B B B B B
B B B B B B B
B B B B B B B
B B B B B B B
Dyad
A A A A A A B B B B B B
A A A A A A B B B B B B
A A A A A A B B B B B B
A A A A A A B B B B B B
Dyad pairwise
A A A A A A B B B B B B
A A A A A A B B B B B B
A A A A A A B B B B B B
A A A A A A B B B B B B
B B B B B B A A A A A A
B B B B B B A A A A A A
B B B B B B A A A A A A
B B B B B B A A A A A A

For the variables that have similar names among the two dyad members (e.g., PA1), we will add a suffix. For instance, partner 1 will have the suffix “_p1” (e.g., ‘PA1_1’), and partner 2 will have “_p2” (e.g., ‘PA1_2’). In both cases, we propose three methods in function of if:

  • dyads’ members are distinguishable (e.g., teacher and student) and a variable identifies a member from the other. In our example, ‘role’ makes dyads distinguishable.
  • dyads are indistinguishable (e.g., friends) and the dyads’ members are laid in the dataframe in a consistent way. The first partner will be on the left and the second on the right of the dataframe.
  • dyads are indistinguishable (e.g., friends) and the dyads’ members are randomly laid in the dataframe, meaning that positions (i.e., right or left) are randomly defined.
Warning

When merging datasets, it is crucial to ensure that the key variables specified in the ‘by’ argument accurately establish correspondence between partner observations within dyads. We can use the ‘dyad’ and ‘obsno’ variables to do so. However, regarding the second variable, if partners within a dyad did not start at the same beep, it may lead to a divergence in partner observations. Here, instead of the ‘obsno’ variable, we prefer using the ‘daycum_dyad’ and ‘beepno’ variables (see Create time variables section). Note that the later is created based on the former one.

Dyad format

The dyad format is a format where partners’ observations are presented only once and displayed side by side.

# Split partners
df1 = data %>% filter(role == 1)
df2 = data %>% filter(role == 2)

# Join side by side
data_dyad = df1 %>% 
    left_join(df2, by=c("dyad", "daycum_dyad", "beepno"), suffix = c("_p1", "_p2"))

The outcome dataframe looks like:

  dyad role_p1 obsno_p1 id_p1 age_p1 cond_dyad_p1
1    1       1        1     1     40        condB
2    1       1        2     1     40        condB
3    1       1        3     1     40        condB
4    1       1        4     1     40        condB
5    1       1        5     1     40        condB
6    1       1        6     1     40        condB
         scheduled_p1             sent_p1
1 2018-10-17 08:00:08 2018-10-17 08:00:11
2 2018-10-17 09:00:01 2018-10-17 09:00:22
3 2018-10-17 09:59:56 2018-10-17 10:00:08
4 2018-10-17 10:59:48 2018-10-17 10:59:52
5 2018-10-17 12:00:12 2018-10-17 12:00:15
6 2018-10-18 07:59:47 2018-10-18 08:00:08
             start_p1              end_p1 contact_p1 PA1_p1
1                <NA>                <NA>         NA     NA
2                <NA>                <NA>         NA     NA
3                <NA>                <NA>         NA     NA
4 2018-10-17 11:00:12 2018-10-17 11:03:01          0      1
5                <NA>                <NA>         NA     NA
6                <NA>                <NA>         NA     NA
  PA2_p1 PA3_p1 NA1_p1 NA2_p1 NA3_p1 location_p1 daycum_p1
1     NA     NA     NA     NA     NA        <NA>    1 days
2     NA     NA     NA     NA     NA        <NA>    1 days
3     NA     NA     NA     NA     NA        <NA>    1 days
4     11     25     10     16     28           A    1 days
5     NA     NA     NA     NA     NA        <NA>    1 days
6     NA     NA     NA     NA     NA        <NA>    2 days
  valid_var_p1 valid_p1 daycum_dyad beepno role_p2 obsno_p2
1            1        0      1 days      1       2        1
2            1        0      1 days      2       2        2
3            1        0      1 days      3       2        3
4            7        1      1 days      4       2        4
5            1        0      1 days      5       2        5
6            1        0      2 days      1       2        6
  id_p2 age_p2 cond_dyad_p2        scheduled_p2
1     2     42        condB 2018-10-17 07:59:58
2     2     42        condB 2018-10-17 08:59:50
3     2     42        condB 2018-10-17 09:59:55
4     2     42        condB 2018-10-17 11:00:01
5     2     42        condB 2018-10-17 11:59:58
6     2     42        condB 2018-10-18 08:00:07
              sent_p2            start_p2
1 2018-10-17 08:00:15 2018-10-17 08:00:21
2 2018-10-17 08:59:53 2018-10-17 09:00:22
3 2018-10-17 10:00:16 2018-10-17 10:00:37
4 2018-10-17 11:00:05 2018-10-17 11:00:15
5 2018-10-17 12:00:01 2018-10-17 12:00:23
6 2018-10-18 08:00:09 2018-10-18 08:00:44
               end_p2 contact_p2 PA1_p2 PA2_p2 PA3_p2
1 2018-10-17 08:04:32          0     32      3     10
2 2018-10-17 09:03:08          0     23      4     14
3 2018-10-17 10:02:42          0     17      7     19
4 2018-10-17 11:03:07          0     24     12     16
5 2018-10-17 12:03:12          0     32      6     14
6 2018-10-18 08:03:49          0     26      1     18
  NA1_p2 NA2_p2 NA3_p2 location_p2 daycum_p2 valid_var_p2
1      5     20      1           C    1 days            7
2      4      4     11           C    1 days            7
3     14     34     20           B    1 days            7
4     24      3     28           A    1 days            7
5     20      1     35           E    1 days            7
6     17      1     42           D    2 days            7
  valid_p2
1        1
2        1
3        1
4        1
5        1
6        1

Dyad pairwise format

The dyad pairwise format is a format where partners’ observations are presented side by side, but each partner’s observations are displayed twice (once as the first partner and once as the second partner).

# Split partners
df1 = data %>% filter(role == 1)
df2 = data %>% filter(role == 2)

# Join side by side
df1_merged = df1 %>% 
    left_join(df2, by=c("dyad", "daycum_dyad", "beepno"), suffix = c("_p1", "_p2"))
df2_merged = df2 %>% 
    left_join(df1, by=c("dyad", "daycum_dyad", "beepno"), suffix = c("_p1", "_p2"))

# Merge
data_dyad = rbind(df1_merged, df2_merged)

The outcome dataframe looks like:

  dyad role_p1 obsno_p1 id_p1 age_p1 cond_dyad_p1
1    1       1        1     1     40        condB
2    1       1        2     1     40        condB
3    1       1        3     1     40        condB
4    1       1        4     1     40        condB
5    1       1        5     1     40        condB
6    1       1        6     1     40        condB
         scheduled_p1             sent_p1
1 2018-10-17 08:00:08 2018-10-17 08:00:11
2 2018-10-17 09:00:01 2018-10-17 09:00:22
3 2018-10-17 09:59:56 2018-10-17 10:00:08
4 2018-10-17 10:59:48 2018-10-17 10:59:52
5 2018-10-17 12:00:12 2018-10-17 12:00:15
6 2018-10-18 07:59:47 2018-10-18 08:00:08
             start_p1              end_p1 contact_p1 PA1_p1
1                <NA>                <NA>         NA     NA
2                <NA>                <NA>         NA     NA
3                <NA>                <NA>         NA     NA
4 2018-10-17 11:00:12 2018-10-17 11:03:01          0      1
5                <NA>                <NA>         NA     NA
6                <NA>                <NA>         NA     NA
  PA2_p1 PA3_p1 NA1_p1 NA2_p1 NA3_p1 location_p1 daycum_p1
1     NA     NA     NA     NA     NA        <NA>    1 days
2     NA     NA     NA     NA     NA        <NA>    1 days
3     NA     NA     NA     NA     NA        <NA>    1 days
4     11     25     10     16     28           A    1 days
5     NA     NA     NA     NA     NA        <NA>    1 days
6     NA     NA     NA     NA     NA        <NA>    2 days
  valid_var_p1 valid_p1 daycum_dyad beepno role_p2 obsno_p2
1            1        0      1 days      1       2        1
2            1        0      1 days      2       2        2
3            1        0      1 days      3       2        3
4            7        1      1 days      4       2        4
5            1        0      1 days      5       2        5
6            1        0      2 days      1       2        6
  id_p2 age_p2 cond_dyad_p2        scheduled_p2
1     2     42        condB 2018-10-17 07:59:58
2     2     42        condB 2018-10-17 08:59:50
3     2     42        condB 2018-10-17 09:59:55
4     2     42        condB 2018-10-17 11:00:01
5     2     42        condB 2018-10-17 11:59:58
6     2     42        condB 2018-10-18 08:00:07
              sent_p2            start_p2
1 2018-10-17 08:00:15 2018-10-17 08:00:21
2 2018-10-17 08:59:53 2018-10-17 09:00:22
3 2018-10-17 10:00:16 2018-10-17 10:00:37
4 2018-10-17 11:00:05 2018-10-17 11:00:15
5 2018-10-17 12:00:01 2018-10-17 12:00:23
6 2018-10-18 08:00:09 2018-10-18 08:00:44
               end_p2 contact_p2 PA1_p2 PA2_p2 PA3_p2
1 2018-10-17 08:04:32          0     32      3     10
2 2018-10-17 09:03:08          0     23      4     14
3 2018-10-17 10:02:42          0     17      7     19
4 2018-10-17 11:03:07          0     24     12     16
5 2018-10-17 12:03:12          0     32      6     14
6 2018-10-18 08:03:49          0     26      1     18
  NA1_p2 NA2_p2 NA3_p2 location_p2 daycum_p2 valid_var_p2
1      5     20      1           C    1 days            7
2      4      4     11           C    1 days            7
3     14     34     20           B    1 days            7
4     24      3     28           A    1 days            7
5     20      1     35           E    1 days            7
6     17      1     42           D    2 days            7
  valid_p2
1        1
2        1
3        1
4        1
5        1
6        1
Warning

The variable ‘obsno’ is not always the most suitable choice for merging dyads’ observations. There are instances where the combination of ‘daycum_dyad’ and ‘beepno’ is more appropriate.

Furthermore, it is crucial to verify that the time interval between partners’ responses adheres to the expected distribution, as described in the dyad time interval section. Specifically, it is important to identify and examine any outliers. Outliers can be when there is a time gap of more than 30 minutes between partner observations that are merged together (unexpected if partners are supposed to answer within an interval of less than 30 minutes). Such outliers may indicate issues with data collection or errors in data manipulation.

# # Gather partners' id number and dyad number
# nested_list = lapply(split(data$id, data$dyad), unique)

# # Split partners
# id_list1 = sapply(nested_list, function(x) x[1])
# id_list2 = sapply(nested_list, function(x) x[2])
# data1 = data[data$id %in% id_list1,]
# data2 = data[data$id %in% id_list2,]

# # Join side by side partners' data
# data1 = data1 %>%
#     left_join(data2, by=c("dyad", "daycum_dyad", "beepno"), suffix = c("_1", "_2"))

# data2 = data2 %>%
#     left_join(data1, by=c("dyad", "daycum_dyad", "beepno"), suffix = c("_1", "_2"))

# data_pairdyad = rbind(data1, data2)