Simulate Normal Population Scores for Multiple Items — calculate

This function allows you to create normal populations for data that would include repeated measures items. Additionally, the data can be rounded and/or truncated to ensure it matches a target scale - for example, a 1-7 type rating scale.

Usage

calculate_proportion(samples, cutoff, grouping_items = NULL, score)

Arguments

samples: The bootstrapped samples from your population
cutoff: The cutoff score for an item to be well measured from the standard errors of your items
grouping_items: The names of columns to group your data by for the cutoff calculation, usually this column is the item column
score: The column of the score you wish to calculate for your cutoff score SE

Value

summary_DF: Summary of the proportion of items below the standard error cutoff score.

Examples

# step 1 create data like what I think I'll get or use your own
pops <- simulate_population(mu = 4, mu_sigma = .2, sigma = 2,
                            sigma_sigma = .2, number_items = 30, number_scores = 20,
                            smallest_sigma = .02, min_score = 1, max_score = 7, digits = 0)

# step 2 calculate our cut off score
cutoff <- calculate_cutoff(population = pops,
 grouping_items = "item",
 score = "score",
 minimum = 1,
 maximum = 7)

cutoff$se_items
#>  [1] 0.3831998 0.4039281 0.4405439 0.4604060 0.3015312 0.4638228 0.4723959
#>  [8] 0.4004931 0.3332456 0.4861232 0.3574397 0.3900067 0.3991768 0.3423987
#> [15] 0.3925289 0.3515005 0.3634629 0.4806793 0.3661679 0.5175855 0.4436689
#> [22] 0.4259664 0.4070109 0.3913539 0.3938675 0.3620119 0.3439324 0.3032716
#> [29] 0.4222434 0.4638228
cutoff$sd_items
#> [1] 0.05510047
cutoff$cutoff
#>       40% 
#> 0.3908151 
cutoff$prop_var
#> [1] 0.01836682

# step 3 simulate samples
samples <- simulate_samples(start = 20, stop = 100,
 increase = 5, population = pops,
 replace = TRUE, grouping_items = "item")

# step 4 and 5
proportion_summary <- calculate_proportion(samples = samples,
 cutoff = cutoff$cutoff,
 grouping_items = "item",
 score = "score")
proportion_summary
#> # A tibble: 17 × 2
#>    sample_size percent_below
#>          <dbl>         <dbl>
#>  1          20         0.529
#>  2          25         0.747
#>  3          30         0.903
#>  4          35         0.98 
#>  5          40         0.999
#>  6          45         1    
#>  7          50         1    
#>  8          55         1    
#>  9          60         1    
#> 10          65         1    
#> 11          70         1    
#> 12          75         1    
#> 13          80         1    
#> 14          85         1    
#> 15          90         1    
#> 16          95         1    
#> 17         100         1