R语言秒转分钟实现问题求助（骑行时长分析场景）

我是R语言新手，正在做骑行时长分析，目前得到的骑行时长结果都是秒单位，不方便做描述性分析和后续可视化，恳请帮忙解决秒转分钟的问题。

现有统计代码及结果 #

我用以下代码计算骑行时长的统计指标：

trip_stats <- cyclistic_df %>%
  group_by(member_casual) %>% 
  summarise(average_ride_length = round((mean(ride_length), 2),  # average ride length (total ride time / trips)
            median_ride_length = round(median(ride_length), 2),  # median ride length
            min_ride_length = round(min(ride_length), 2),  # minimum ride length
            max_ride_length = round(max(ride_length), 2))  # maximum ride length
head(trip_stats)

运行结果：

# A tibble: 2 × 5
  member_casual average_ride_length median_ride_length min_ride_length max_ride_length
  <chr>         <drtn>              <drtn>             <drtn>          <drtn>         
1 casual        22.72 secs          785 secs           1 secs          1922127 secs   
2 member        12.19 secs          525 secs           1 secs            89872 secs

单独计算平均时长的代码：

# Average ride length (ride_length):
ride_lengt_avg <- round(mean(cyclistic_df$ride_length), 2)
print(ride_lengt_avg)

运行结果：

Time difference of 977.28 secs

尝试过的方法及问题 #

我试过as_hms、format("%M:%S")、minutes()等方法，结果还是秒单位。比如我用以下代码提取小时列是成功的：

# Format time as HH:MM:SS:
cyclistic_df$time <- format(as.Date(cyclistic_df$date), "%H:%M:%S")

# Create new column for time:
cyclistic_df$time <- as_hms((cyclistic_df$started_at))

# Create new column for hour:
cyclistic_df$hour <- hour(cyclistic_df$time)

运行结果：

0      1      2      3      4      5      6      7      8      9     10     11     12     13     14     15     16     17
  casual  32053  20813  12246   6763   4515   8707  23390  40346  55066  56190  72233  93437 109905 114531 121187 134839 153614 170534
  member  25582  15645   8736   5268   6097  26049  81660 151664 180585 120035 109540 129717 149403 147804 148800 183330 249039 297724
        
             18     19     20     21     22     23
  casual 148452 111916  81388  69348  61476  44808
  member 233303 165938 115161  88668  65505  41332

但没法用类似方法把骑行时长转成分钟，直接除以60的话，结果还是会保留"secs"标识，容易造成误导。

已添加的列及疑问 #

我已经给数据集添加了以下列：

# Default format is yyyy-mm-dd, use start date:
cyclistic_df$date <- as.Date(cyclistic_df$started_at)  # the default format is yyyy-mm-dd

# Create column for year:
cyclistic_df$year <- format(as.Date(cyclistic_df$date), "%Y")

# Create column for month:
cyclistic_df$month <- format(as.Date(cyclistic_df$date), "%m")

# Create column for day:
cyclistic_df$day <- format(as.Date(cyclistic_df$date), "%d")

# Calculate the day of the week:
cyclistic_df$day_of_week <- wday(cyclistic_df$started_at)

# Create column for day of week:
cyclistic_df$day_of_week <- format(as.Date(cyclistic_df$date), "%A")  # wday(cyclistic_df$started_at, label = T, abbr = T)

# Format time as HH:MM:SS:
cyclistic_df$time <- format(as.Date(cyclistic_df$date), "%H:%M:%S")

# Create new column for time:
cyclistic_df$time <- as_hms((cyclistic_df$started_at))

# Create new column for hour:
cyclistic_df$hour <- hour(cyclistic_df$time)

# Calculate & Create ride length column by subtracting ended_at time from started_at time and converted it to minutes:
cyclistic_df$ride_length <- as_hms(difftime(cyclistic_df$ended_at, cyclistic_df$started_at))

我觉得year、month、day、hour、ride_length这些列应该设为数值类型，避免后续重复转换，是不是最好不要用format( , "%__")这种方法？

内容的提问来源于stack exchange，提问作者prolabrus