understanding online fraud victimisation in...

Understanding online fraud victimisation in Australia

15th International Symposium of World Society of Victimology Dr. Monique Mann, Mr. Anthony Morgan and Dr. Marcus Smith

Australian Institute of Criminology

Online Fraud

• Four defining elements of online fraud (Cross, Smith & Richards, 2014)

1.Via the internet 2.Dishonest invitation, request, notification or offer 3.Provide personal information or money 4.Financial or non-financial loss or impact of some kind

$1.4 billion lost to personal fraud

1.2 million Australian victims

Source

Jurisdiction

Sample size

Key findings

Reyns (2013)

United Kingdom 5,985

Internet banking, IM, emailing, males, older persons, higher incomes, perceive risk have increased likelihood of being a victim of identity theft.

Leukfeldt (2014)

Europe (Netherlands) 10,316

Demographic and financial characteristics did not impact the likelihood of phishing victimisation. Antivirus software had no impact on victimisation.

Reisig, Pratt & Holtfreter (2009)

United States (Florida) 573

Demographics (socially vulnerable, lower socio-economic status) influence risk perception. Individuals with higher perceived risk spend less time online and make fewer online purchases

Source

Jurisdiction

Sample size

Key findings

Reyns (2013)

United Kingdom 5,985

Internet banking, IM, emailing, males, older persons, higher incomes, perceive risk have increased likelihood of being a victim of identity theft.

Leukfeldt (2014)

Europe (Netherlands) 10,316

Demographic and financial characteristics did not impact the likelihood of phishing victimisation. Antivirus software had no impact on victimisation.

Reisig, Pratt & Holtfreter (2009)

United States (Florida) 573

Demographics (socially vulnerable, lower socio-economic status) influence risk perception. Individuals with higher perceived risk spend less time online and make fewer online purchases

Source

Jurisdiction

Sample size

Key findings

Reyns (2013)

United Kingdom 5,985

Internet banking, IM, emailing, males, older persons, higher incomes, perceive risk have increased likelihood of being a victim of identity theft.

Leukfeldt (2014)

Europe (Netherlands) 10,316

Demographic and financial characteristics did not impact the likelihood of phishing victimisation. Antivirus software had no impact on victimisation.

Reisig, Pratt & Holtfreter (2009)

United States (Florida) 573

Demographics (socially vulnerable, lower socio-economic status) influence risk perception. Individuals with higher perceived risk spend less time online and make fewer online purchases

Source

Jurisdiction

Sample size

Key findings

Reyns (2013)

United Kingdom 5,985

Internet banking, IM, emailing, males, older persons, higher incomes, perceive risk have increased likelihood of being a victim of identity theft.

Leukfeldt (2014)

Europe (Netherlands) 10,316

Demographic and financial characteristics did not impact the likelihood of phishing victimisation. Antivirus software had no impact on victimisation.

Reisig, Pratt & Holtfreter (2009)

United States (Florida) 573

Demographics (socially vulnerable, lower socio-economic status) influence risk perception. Individuals with higher perceived risk spend less time online and make fewer online purchases

Routine Activity Theory (Cohen & Felson, 1979)

Suitable target

Motivated offender Absence of a

capable guardian

ONLINE FRAUD VICTIMISATION

Suitable target

Motivated offender Absence of a

capable guardian

ONLINE FRAUD VICTIMISATION

What is the relationship between victimisation and:

• Demographics • Online activities

• Perceived risk • Prevention behaviour

Method and Analytic Approach

Online fraud victimisation (n=1,786)

n %

Victim 345 19.32

Not Victim 1,441 80.68

Category Independent Variables

Demographics Gender

Age

Education

Income

Online activities Social media use

Tablet use

Online shopping

Preventative behaviour Passwords

Security measures including anti-virus

Financial protection

Spam filters

Interact with known persons only

Category Independent Variables

Demographics Gender

Age

Education

Income

Online activities Social media use

Tablet use

Online shopping

Preventative behaviour Passwords

Security measures including anti-virus

Financial protection

Spam filters

Interact with known persons only

Category Independent Variables

Demographics Gender

Age

Education

Income

Online activities Social media use

Tablet use

Online shopping

Preventative behaviour Passwords

Security measures including anti-virus

Financial protection

Spam filters

Interact with known persons only

Relationship between education and victimisation

13

19

26 28

14

18

22

29

22

9

0

5

10

15

20

25

30

35

40

45

50

Less than Year 12 orSecondary School

Year 12 or equivalent Vocational qualification Bachelor degree Postgraduate degree

%

Victim

Not victim

χ2 (4, n= 1,786) = 18.03, p.<.001, Cramer’s V= 0.1 (small effect size)

Relationship between income and victimisation

11

25

33

17

4

11

17

24

29

16

2

12

0

5

10

15

20

25

30

35

40

45

50

$0-18,200 $18,201-37,000 $37,001-80,000 $80,001-$180,000 $180,001 and over I'd rather not say

%

Victim

Not victim

χ2 (5, n= 1,786) = 11.40, p.<.05, Cramer’s V= 0.08 (small effect size)

Relationship between online shopping and victimisation

2

10

17 20

31

20

6

23 20

23 21

6

0

5

10

15

20

25

30

35

40

45

50

Never Less often Every 2-3 months Monthly 2-3 times a month Weekly or moreoften

%

Victim

Not victim

χ2 (5, n= 1,786) = 80.40, p.<.000, Cramer’s V= 0.21 (medium effect size)

Relationship between risk perception and victimisation

41

59

30

70

0

10

20

30

40

50

60

70

80

90

100

Perceived risk No risk

%

VictimNot victim

χ2 (1, n=1,703) = 15.90, p.<.000, Cramer’s V= 0.1 (small effect size)

Logistic Regression Predicting Victimisation

Predictors Odds Ratio (Ѱ) 95%CIѰ p<

Income

$80,001-$180,000 (vs $37,001-$80,000) 0.68 0.47-1.00 .05

Online shopping

Every 2-3 months (vs 2-3 times a month) 0.56 0.39-0.80 .01

Less often than 2-3 months (vs 2-3 times a month) 0.35 0.23-0.55 .001

Never (vs 2-3 times a month) 0.34 0.16-0.75 .01

Weekly or more often (vs 2-3 times a month) 1.86 1.26-2.74 .01

Risk perception (vs no risk perception) 1.40 1.07-1.81 .05

Model: χ2 (df=18, n= 1,703) = 102.34, p<.05, AUC = 0.67, Pseudo R2 = 0.06, Cox & Snell R2=0.06, Nagelkerke R2= 0.09

Risk factors: 1. Income

• $37,001-$80,000 2. Frequent online shoppers

• Weekly or more often • 2-3 times per month • Every 2-3 months • Less often than 2-3 months • Never

3. Risk perception

R

ISK

Key findings

• Complicated risk profile, divergences with traditional victim profile

• Analysis has gone a little further, but only explains a small amount of risk

• Need to take a more nuanced approach to understanding online fraud

Relationship between risk perception and prevention behaviour

49

82 78

56

86 84

0102030405060708090

100

Use spam filters Use financialprotection

Interact knownpersons only

%

Risk

No Risk

The way forward

• Examine what influences the likelihood of victimisation for other online crimes (e.g. attacks on computer system and virus), and differences for sub-categories of online fraud (e.g. romance scams versus work from home scams)

• Investigate protective factors and impact of preventative strategies

for specific sub-categories of online fraud • Examine characteristics and predictors of online repeat victims,

limited literature that examines repeat victimisation in online context

QUESTIONS?

Contact: Monique.Mann@aic.gov.au