Datadive4.Rmd
2024-02-04
data <- read.csv ("C:\\Users\\varsh\\OneDrive\\Desktop\\Gitstuff\\age_gaps.CSV")set.seed(42)
for (i in 1:5) {
sample_size <- round(0.5 * nrow(data))
random_indices <- sample(1:nrow(data), size = sample_size, replace = TRUE)
sample_df <- data[random_indices, ]
assign(paste0("df_", i), sample_df)
}print("First 5 rows of df_1:")## [1] "First 5 rows of df_1:"head(df_1)## movie_name release_year director age_difference
## 561 Y tu mamá también 2001 Alfonso Cuarón 9
## 321 From Russia with Love 1963 Terence Young 14
## 1098 Slackers 2002 Dewey Nicks 1
## 634 Click 2006 Frank Coraci 7
## 49 Café Society 2016 Woody Allen 28
## 1152 The Lake House 2006 Alejandro Agresti 0
## couple_number actor_1_name actor_2_name character_1_gender
## 561 1 Maribel Verdú Diego Luna woman
## 321 1 Sean Connery Aliza Gur man
## 1098 1 Devon Sawa Jaime King man
## 634 1 Adam Sandler Kate Beckinsale man
## 49 1 Steve Carell Kristen Stewart man
## 1152 1 Sandra Bullock Keanu Reeves woman
## character_2_gender actor_1_birthdate actor_2_birthdate actor_1_age
## 561 man 02-10-1970 29-12-1979 31
## 321 woman 25-08-1930 01-04-1944 33
## 1098 woman 07-09-1978 23-04-1979 24
## 634 woman 09-09-1966 26-07-1973 40
## 49 woman 16-08-1962 09-04-1990 54
## 1152 man 26-07-1964 02-09-1964 42
## actor_2_age
## 561 22
## 321 19
## 1098 23
## 634 33
## 49 26
## 1152 42print("First 5 rows of df_2:")## [1] "First 5 rows of df_2:"head(df_2)## movie_name release_year director
## 298 Once Upon a Time in Hollywood 2019 Quentin Tarantino
## 17 Shopgirl 2005 Anand Tucker
## 482 Husbands and Wives 1992 Woody Allen
## 532 It's Complicated 2009 Nancy Meyers
## 76 I Married a Witch 1942 René Clair
## 1112 The Theory of Everything 2014 James Marsh
## age_difference couple_number actor_1_name actor_2_name
## 298 15 1 Leonardo DiCaprio Lorenza Izzo
## 17 34 1 Steve Martin Claire Danes
## 482 10 3 Woody Allen Mia Farrow
## 532 9 2 Meryl Streep Alec Baldwin
## 76 25 1 Fredric March Veronica Lake
## 1112 1 1 Eddie Redmayne Felicity Jones
## character_1_gender character_2_gender actor_1_birthdate actor_2_birthdate
## 298 man woman 11-11-1974 19-09-1989
## 17 man woman 14-08-1945 12-04-1979
## 482 man woman 01-12-1935 09-02-1945
## 532 woman man 22-06-1949 03-04-1958
## 76 man woman 1897-08-31 14-11-1922
## 1112 man woman 26-01-1982 17-10-1983
## actor_1_age actor_2_age
## 298 45 30
## 17 60 26
## 482 57 47
## 532 60 51
## 76 45 20
## 1112 32 31print("First 5 rows of df_3:")## [1] "First 5 rows of df_3:"head(df_3)## movie_name release_year director
## 956 Trainwreck 2015 Judd Apatow
## 474 EDtv 1999 Ron Howard
## 383 Casino Royale 2006 Martin Campbell
## 353 How Stella Got Her Groove Back 1998 Kevin Rodney Sullivan
## 383.1 Casino Royale 2006 Martin Campbell
## 155 Get Smart 2008 Peter Segal
## age_difference couple_number actor_1_name actor_2_name
## 956 3 2 Bill Hader Amy Schumer
## 474 10 1 Woody Harrelson Jenna Elfman
## 383 12 1 Daniel Craig Eva Green
## 353 13 1 Angela Bassett Taye Diggs
## 383.1 12 1 Daniel Craig Eva Green
## 155 20 1 Steve Carell Anne Hathaway
## character_1_gender character_2_gender actor_1_birthdate actor_2_birthdate
## 956 man woman 07-06-1978 01-06-1981
## 474 man woman 23-07-1961 30-09-1971
## 383 man woman 02-03-1968 06-07-1980
## 353 woman man 16-08-1958 02-01-1971
## 383.1 man woman 02-03-1968 06-07-1980
## 155 man woman 16-08-1962 12-11-1982
## actor_1_age actor_2_age
## 956 37 34
## 474 38 28
## 383 38 26
## 353 40 27
## 383.1 38 26
## 155 46 26print("First 5 rows of df_4:")## [1] "First 5 rows of df_4:"head(df_4)## movie_name release_year director age_difference
## 1153 Tolkien 2019 Dome Karukoski 0
## 506 The Time Traveler's Wife 2009 Robert Schwentke 10
## 470 Daddy's Home 2015 Sean Anders 10
## 1149 The Crush 1993 Alan Shapiro 0
## 617 Tommy Boy 1995 Peter Segal 8
## 932 Savages 2007 undefined 3
## couple_number actor_1_name actor_2_name character_1_gender
## 1153 1 Lily Collins Nicholas Hoult woman
## 506 1 Eric Bana Rachel McAdams man
## 470 1 Mark Wahlberg Alessandra Ambrosio man
## 1149 2 Jennifer Rubin Cary Elwes woman
## 617 2 Bo Derek Rob Lowe woman
## 932 2 Blake Lively Aaron Taylor-Johnson woman
## character_2_gender actor_1_birthdate actor_2_birthdate actor_1_age
## 1153 man 18-03-1989 07-12-1989 30
## 506 woman 09-08-1968 17-11-1978 41
## 470 woman 05-06-1971 11-04-1981 44
## 1149 man 03-04-1962 26-10-1962 31
## 617 man 20-11-1956 17-03-1964 39
## 932 man 25-08-1987 13-06-1990 20
## actor_2_age
## 1153 30
## 506 31
## 470 34
## 1149 31
## 617 31
## 932 17print("First 5 rows of df_5:")## [1] "First 5 rows of df_5:"head(df_5)## movie_name release_year director age_difference
## 316 Because I Said So 2007 Michael Lehmann 14
## 294 Home Again 2017 Hallie Meyers-Shyer 15
## 276 The Family Stone 2007 Thomas Bezucha 16
## 760 Days of Thunder 1990 Tony Scott 5
## 1118 Troy 2004 Wolfgang Petersen 1
## 699 Chasing Liberty 2004 Andy Cadiff 6
## couple_number actor_1_name actor_2_name character_1_gender
## 316 1 Tom Everett Scott Mandy Moore man
## 294 1 Reese Witherspoon Pico Alexander woman
## 276 1 Dermot Mulroney Claire Danes man
## 760 1 Tom Cruise Nicole Kidman man
## 1118 2 Diane Kruger Orlando Bloom woman
## 699 1 Matthew Goode Mandy Moore man
## character_2_gender actor_1_birthdate actor_2_birthdate actor_1_age
## 316 woman 07-09-1970 10-04-1984 37
## 294 man 21-03-1976 03-06-1991 41
## 276 woman 31-10-1963 12-04-1979 44
## 760 woman 03-07-1962 20-06-1967 28
## 1118 man 15-07-1976 13-01-1977 28
## 699 woman 03-04-1978 10-04-1984 26
## actor_2_age
## 316 23
## 294 26
## 276 28
## 760 23
## 1118 27
## 699 20Insight Gained:
This code generates five random subsamples (df_1 through df_5) from the original dataset. The key insights obtained concern the range and unpredictability generated by the sampling procedure.
By comparing the top five rows of each subsample, we can see whether the selected observations differ between the samples.
Significance:
The differences between the first five rows of each subsample demonstrate how random sampling affects data composition.
Understanding how distinct subsamples can vary helps in determining the representativeness of specific samples and their potential impact on results.
Further Questions:
1. Do specific patterns or trends show or disappear consistently across subsamples?
2. Are some observations regularly overrepresented or underrepresented in subsamples?
3. How does variability in subsamples affect summary statistics like means, standard deviations, and other critical metrics?
analyze_subsample <- function(subsample, subsample_number) {
cat("Analysis for Subsample", subsample_number, "\n")
summary_stats <- summary(subsample)
print(summary_stats)
anomalies <- as.data.frame(table(subsample$character_1_gender, subsample$character_2_gender))
colnames(anomalies) <- c("character_1_gender", "character_2_gender", "Count")
cat("Anomalies in character_1_gender and character_2_gender:\n")
print(anomalies)
column_to_simulate <- "age_difference"
num_simulations <- 1000
simulated_means <- replicate(num_simulations, mean(sample(subsample[[column_to_simulate]], replace = TRUE)))
cat("Monte Carlo Simulation for", column_to_simulate, "Mean:\n")
hist(simulated_means, main = paste("Distribution of Simulated Means for", column_to_simulate))
}list_dfs = list(df_1, df_2, df_3, df_4, df_5)
for (i in 1:5){
analyze_subsample(list_dfs[[i]], i)
}## Analysis for Subsample 1
## movie_name release_year director age_difference
## Length:578 Min. :1939 Length:578 Min. : 0.00
## Class :character 1st Qu.:1996 Class :character 1st Qu.: 3.00
## Mode :character Median :2003 Mode :character Median : 8.00
## Mean :2001 Mean :10.04
## 3rd Qu.:2012 3rd Qu.:15.00
## Max. :2022 Max. :50.00
## couple_number actor_1_name actor_2_name character_1_gender
## Min. :1.000 Length:578 Length:578 Length:578
## 1st Qu.:1.000 Class :character Class :character Class :character
## Median :1.000 Mode :character Mode :character Mode :character
## Mean :1.382
## 3rd Qu.:2.000
## Max. :6.000
## character_2_gender actor_1_birthdate actor_2_birthdate actor_1_age
## Length:578 Length:578 Length:578 Min. :18.00
## Class :character Class :character Class :character 1st Qu.:32.25
## Mode :character Mode :character Mode :character Median :38.00
## Mean :39.92
## 3rd Qu.:46.00
## Max. :79.00
## actor_2_age
## Min. :17.00
## 1st Qu.:25.00
## Median :29.00
## Mean :29.88
## 3rd Qu.:34.00
## Max. :61.00
## Anomalies in character_1_gender and character_2_gender:
## character_1_gender character_2_gender Count
## 1 man man 8
## 2 woman man 107
## 3 man woman 459
## 4 woman woman 4
## Monte Carlo Simulation for age_difference Mean:
## Analysis for Subsample 2
## movie_name release_year director age_difference
## Length:578 Min. :1935 Length:578 Min. : 0.0
## Class :character 1st Qu.:1998 Class :character 1st Qu.: 3.0
## Mode :character Median :2005 Mode :character Median : 8.0
## Mean :2002 Mean :10.6
## 3rd Qu.:2012 3rd Qu.:15.0
## Max. :2022 Max. :52.0
## couple_number actor_1_name actor_2_name character_1_gender
## Min. :1.0 Length:578 Length:578 Length:578
## 1st Qu.:1.0 Class :character Class :character Class :character
## Median :1.0 Mode :character Mode :character Mode :character
## Mean :1.4
## 3rd Qu.:2.0
## Max. :6.0
## character_2_gender actor_1_birthdate actor_2_birthdate actor_1_age
## Length:578 Length:578 Length:578 Min. :18.00
## Class :character Class :character Class :character 1st Qu.:32.25
## Mode :character Mode :character Mode :character Median :39.00
## Mean :40.84
## 3rd Qu.:48.00
## Max. :79.00
## actor_2_age
## Min. :17.00
## 1st Qu.:24.00
## Median :29.00
## Mean :30.24
## 3rd Qu.:34.00
## Max. :60.00
## Anomalies in character_1_gender and character_2_gender:
## character_1_gender character_2_gender Count
## 1 man man 11
## 2 woman man 113
## 3 man woman 446
## 4 woman woman 8
## Monte Carlo Simulation for age_difference Mean:
## Analysis for Subsample 3
## movie_name release_year director age_difference
## Length:578 Min. :1936 Length:578 Min. : 0.0
## Class :character 1st Qu.:1996 Class :character 1st Qu.: 4.0
## Mode :character Median :2004 Mode :character Median : 9.0
## Mean :2000 Mean :10.5
## 3rd Qu.:2011 3rd Qu.:15.0
## Max. :2021 Max. :52.0
## couple_number actor_1_name actor_2_name character_1_gender
## Min. :1.000 Length:578 Length:578 Length:578
## 1st Qu.:1.000 Class :character Class :character Class :character
## Median :1.000 Mode :character Mode :character Mode :character
## Mean :1.394
## 3rd Qu.:2.000
## Max. :6.000
## character_2_gender actor_1_birthdate actor_2_birthdate actor_1_age
## Length:578 Length:578 Length:578 Min. :19.00
## Class :character Class :character Class :character 1st Qu.:34.00
## Mode :character Mode :character Mode :character Median :39.00
## Mean :40.58
## 3rd Qu.:47.00
## Max. :79.00
## actor_2_age
## Min. :17.00
## 1st Qu.:25.00
## Median :29.00
## Mean :30.08
## 3rd Qu.:34.00
## Max. :58.00
## Anomalies in character_1_gender and character_2_gender:
## character_1_gender character_2_gender Count
## 1 man man 7
## 2 woman man 105
## 3 man woman 463
## 4 woman woman 3
## Monte Carlo Simulation for age_difference Mean:
## Analysis for Subsample 4
## movie_name release_year director age_difference
## Length:578 Min. :1935 Length:578 Min. : 0.00
## Class :character 1st Qu.:1997 Class :character 1st Qu.: 3.00
## Mode :character Median :2004 Mode :character Median : 8.00
## Mean :2001 Mean :10.67
## 3rd Qu.:2011 3rd Qu.:15.00
## Max. :2022 Max. :52.00
## couple_number actor_1_name actor_2_name character_1_gender
## Min. :1.000 Length:578 Length:578 Length:578
## 1st Qu.:1.000 Class :character Class :character Class :character
## Median :1.000 Mode :character Mode :character Mode :character
## Mean :1.407
## 3rd Qu.:2.000
## Max. :7.000
## character_2_gender actor_1_birthdate actor_2_birthdate actor_1_age
## Length:578 Length:578 Length:578 Min. :19.0
## Class :character Class :character Class :character 1st Qu.:34.0
## Mode :character Mode :character Mode :character Median :39.0
## Mean :40.9
## 3rd Qu.:47.0
## Max. :79.0
## actor_2_age
## Min. :17.00
## 1st Qu.:25.00
## Median :29.00
## Mean :30.23
## 3rd Qu.:34.00
## Max. :68.00
## Anomalies in character_1_gender and character_2_gender:
## character_1_gender character_2_gender Count
## 1 man man 5
## 2 woman man 108
## 3 man woman 461
## 4 woman woman 4
## Monte Carlo Simulation for age_difference Mean:
## Analysis for Subsample 5
## movie_name release_year director age_difference
## Length:578 Min. :1935 Length:578 Min. : 0.00
## Class :character 1st Qu.:1995 Class :character 1st Qu.: 4.00
## Mode :character Median :2003 Mode :character Median : 8.50
## Mean :2000 Mean :10.67
## 3rd Qu.:2011 3rd Qu.:16.00
## Max. :2022 Max. :50.00
## couple_number actor_1_name actor_2_name character_1_gender
## Min. :1.000 Length:578 Length:578 Length:578
## 1st Qu.:1.000 Class :character Class :character Class :character
## Median :1.000 Mode :character Mode :character Mode :character
## Mean :1.396
## 3rd Qu.:2.000
## Max. :7.000
## character_2_gender actor_1_birthdate actor_2_birthdate actor_1_age
## Length:578 Length:578 Length:578 Min. :19.0
## Class :character Class :character Class :character 1st Qu.:33.0
## Mode :character Mode :character Mode :character Median :39.0
## Mean :40.4
## 3rd Qu.:47.0
## Max. :79.0
## actor_2_age
## Min. :17.00
## 1st Qu.:24.00
## Median :29.00
## Mean :29.73
## 3rd Qu.:34.00
## Max. :68.00
## Anomalies in character_1_gender and character_2_gender:
## character_1_gender character_2_gender Count
## 1 man man 5
## 2 woman man 96
## 3 man woman 474
## 4 woman woman 3
## Monte Carlo Simulation for age_difference Mean:
Insights Gathered:
The age differences between actors in each subsample vary, with averages ranging from around 10.06 to 10.38.
All subsamples have anomalies in the ‘character_1_gender’ and ‘character_2_gender’ variables.
Also, a Monte Carlo simulation is performed to obtain a distribution of simulated means for the ‘age_difference’ variable, allowing an assessment of the stability and variability of this significant metric.
Significance:
The variation in age differences indicates that particular subsamples may have a higher or lower average age gap between actors.
Understanding these differences is critical for evaluating the dataset’s diversity in age dynamics.
Gender anomalies show inconsistencies or unexpected results in character gender assignments.
Addressing these abnormalities is critical to maintaining accuracy and dependability in later analysis.
Assessing the distribution of simulated means for ‘age_difference’ may help in identifying the metric’s stability over different random samples. The simulation reveals how changes in sample make up may affect the overall mean age difference.
Further Investigation:
Analyze the factors that influence age differences across subsamples. Are some films or filmmakers connected with greater or lesser age differences?
Understand the nature of anomalies and, if necessary, clean up the data. Are there any trends or commonalities among the misassignments?
Examine the stability of mean age differences throughout Monte Carlo simulations. Are there steady trends, or is the metric highly variable?